Individual differences and prejudice

Individual differences and prejudice
Order Description
complete the Right-wing Authoritarianism Scale (https://www.panojohnson.com/automatons/rwa-scale.xhtml (Links to an external site.))
https://www.panojohnson.com/automatons/rwa-scale.xhtml

OR the measure of Social Dominance Orientation (https://en.wikipedia.org/wiki/Social_dominance_orientation#SDO_Scale (Links to an external site.),
https://en.wikipedia.org/wiki/Social_dominance_orientation#SDO_Scale

you need to calculate the score manually though, note the reverse itemsas you believe famous people might (if possible, choose people who are likely to hold extremely different attitudes, such as Rush Limbaugh versus Jon Stewart). Then, discuss the reasons for your choices and relate them to the research on those two measures described in the text.
Chapter 7
*
Individual Differences
and Prejudice
[Sonle people] are so hostile toward so many minOlities,
they seem to be equal opportunity bigots,
-BOB ALTEMEYER (1998, P. 52)
Chapter Outline
Authoritarianism and Social
Dominance Orientation
Authoritarianism
Social Dominance Orientation
Self-Esteem
Self-Enhancement
Self-Protection
Personal Values
Value Orientations
Perceived Value Differences
Social Ideologies
Religion
Political Orientation
Summary
Suggested Reading
Key Terms
Questions for Review and
Discussion
Ate there people who dislike all outgroups equally and so are, in Bob
Altemeyer’s (1998) words quoted above, “equal opportunity bigots” (p, 52)?
That is, are there people who are, as a result of their personalities, belief systems, or
other personal characteristics, especially likely to become prejudiced, and become
prejudiced toward not just Oile group, but toward everyone they see as different
from themselves? Individual difference researchers address these questions by studying
the ways in which people differ from one another and the ways in which these
232
INDIVIDUAL DIFFERENCES AND PREJUDICE 233
personal characteristics are related to other variables such as prejudice. Individual differences
began to become important to the study of prejudice after World War II,
when researchers concluded that factors such as realistic intergroup conflict and
competition (see Chapter 9) could not explain Nazi anti-Semitism and the
Holocaust. “Explanations were therefore sought in the disturbed personality, for it
was hardly conceivable that these actions could be the actions of nonnal men”
(Milner, 1981, p. 106). This search led to the development of one of the first individual
difference theories of prejudice, the theory of the authoritarian personality,
which we discuss shortly. A second reason why researchers believe that individual
differences play a role in prejudice is that researchers have found that people who
score high on prejudice against one group also tend to score high on prejudice
against other groups (for example, Cunningham, Nezlek, & Banaji, 2004). This
similarity of response to different groups suggests that some characteristic of the
person may be a common underlying cause of all the prejudices.
This chapter examines the relationships of selected :individual difference variables
to prejudice. Choosing the variables to discuss was not easy. Researchers have
studied more than 25 individual difference variables :in relation to prejudice
(McFarland, 2001); however, in a set of four studies, Sam McFarland (2001) found
that wo of those variables-authoritarianism and social dom:inance orientationwere
consistently related to prejudice. This chapter begins, therefore, with those
wo variables. The second section of this chapter focuses on the role of the self
in prejudice, both because of its historical importance and because of some recent
theories that COIlllect some self-related variables with prejudice, The third section
examines the role of value systems in prejudice, and the last section focuses on two
important social ideologies related to prejudice—religion and political orientation,
While reading this chapter, it is important to bear wo cautions in mind.
First, research on individual differences is, by its nature, correlational. As we discussed
in Chapter 2, although correlational research can show that two variables
are related to each other, it cannot show that one of those variables causes the
other. Second, the relationships benveen individual difference variables and prejudice
are far from perfect, so a high score on an individual difference variable
that is related to prejudice does not necessarily mean that the person is prejudiced;
other factors can offset the effect of any particular variable, Conversely,
not all people who score low on an individual difference variable related to prejudice
are unprejudiced; other factors can lead the person to be prejudiced.
AUTHORITARIANISM AND SOCIAL
DOMINANCE ORIENTATION
Authoritarianism and social dominance orientation are nvo individual difference
variables closely associated with prejudice (McFarland, 2001). Although authori~’
tarianism was originally conceptualized as a personality trait and social dominance
orientation is frequendy referred to as one, John Duckitt (2001) has pointed out
that they more closely resemble ideologies, sets of attitudes and beliefs that
234 CHAPTER 7
\ predispose people to view the world in certain ways and to respond in ways
l:0nsistent with those viewpoints. For example, from an authoritarian perspective,
the world is a dangerous place, so people high in authoritarianism seek
security by trying to make the world conform to their political and social values.
In a similar vein, people high in social dominance orientation see the world as a
competitive jungle and respond by trying to prevent people and groups they see
as competitors from gaining on them (Duckitt, 2001).
Authoritarianism
Theodor Adorno and !tis colleagues (Adorno, Frenkel-Brunswik, Levinson, &
Sanford, 1950) developed the concept of the authoritarian personality as a
means of explaining the rise of fascism during the 19305. Fascism is a political philosophy
that holds, among other tenets, that those who hold power in a society
know what is best for the society, so people should simply do what their govemrnent
tells them to do. Fascism was quite popular in Europe and the United States
during the 1930s; fascist governments were established in Gernlany, Italy, and
Spain, and fascist movements existed in the United States and Great Britain. It
was the German fascist (or Nazi) government that directed the systematic annihilation
of ethnic and racial minority groups and the mentally and physically handicapped
that is known as the Holocaust. As noted earlier, Adorno and his
colleagues, along with other researchers, began to look for an explanation for
why large numbers of people could become complicit in government-led genocide.
They believed that the scope of the Holocaust meant that it could not be
explained in tenns of intergroup conflict, so the answer must lie within the human
mind. They therefore postulated the existence of what they called the authoritarian
personality, a personality type that is especially susceptible to unthinking
obedience to authority.
[The Authoritarian Personality. Adorno and his colleagues (1950) proposed
that the authoritarian personality was composed of nine characteristic patterns
of thought, five of which are related to prejudice;
1. Conventionalism. Rigid adherence to conventional, middle-class values.
2. Authoritarian submission. Submissive uncritical attitude toward idealized
moral authorities ….
3. Authoritarian aggression. Tendency to be on the lookout for, and to condemn,
reject, and punish people who violate conventional values … .
4. Stereotypy. The … disposition to think in rigid categories … .
5. Projectivity. The disposition to believe that wild and dangerous things go
on in the world; the projection outwards of unconscious emotional
impulses. (p. 228)
Adorno and his colleagues believed that the propensity for rigid adherence
to conventional thinking leads people with authoritarian personalities to view
the world in stereotypical tenus; conventionalisul and authoritarian submission
INDIVIDUAL DIFFERENCES AND PREJUDICE 235
c0111bined with authoritarian aggression leads them to be prejudiced against people
who violate conventional nonns or who are condemned by authority figures;
and projectivity leads them to see their own faults in the targets of their
prejudice.
Adorno and colleagues (1950) used interviews and questionnaires to examine
the relationship of the authoritatian personality to prejudice. One of the questionnaires,
the F-Scale, has become a standard measure of authoritarianism (Meloerg
1993). Adorno and his colleagues found high correlations between authoritarianism
and prejudice against a variety of ethnic groups. Later research using the
F-Scale also found support for a relationship between authoritarianism and ethnic
and racial prejudice (Duckitt, 1994), and scores on the F-Scale also correlate
strongly with attitudes toward lesbians and gay men (Whitley & Lee, 2000).
Thus, there is good evidence that the authoritarian personality, at least as assessed
by the F-Scale, is associated with prejudice.
Despite its early popularity and success, interest in the authoritarian personality
began to decline in the 1960s and 1970s. There were several reasons for this
change. One was a growing disenchantment among psychologists with psychoanalytic
theory on which Adorno and his colleagues (1950) based their theory
and a simultaneous growth in interest in the cognitive underpinnings of prejudice
(Duckitt, 1994). In addition, a number of flaws were found in the F-Scale, which
led to some loss of faith in the original research results; however, subsequent revisions
of the F-Scale have corrected those shortcomings (Christie, 1991). A final
criticism was that although Adorno and his colleagues conceptualized the authoritarian
personality as a characteristic of the political far right-wing, people on the
far left could also show some characteristics of the authoritarian personality, such
as uncritical acceptance of statements made by authority figures and aggression
toward people who do not share their beliefs (Stone & Smith, 1993). This criticism
led to attempts to develop measures of generalized authoritarianism that
would capture both its right- and left-wing aspects, such as Milton Rokeach’s
(1960) Dogmatisnl Scale. However, such attempts have not been very successful;
for example, although the Dogmatism Scale was designed to be politically neutral,
scores on it correlate fairly highly with scores on measures of right-wing
authoritarianism, suggesting that the measures assess similar traits (Altemeyer,
1996). Because of problems such as these, recent research has focused on the
relationship between what is now called right-wing authoritarianism and prejudice.
Right-wing Authoritarianism. Mter languishing during the 1970s, research
on authoritarianism was revived by Bob Altemeyer (1981, 1988, 1996), who replaced
the concept of the authoritarian personality with that of right-wing
authoritarianism (RWA). R WA differs somewhat from the original concept
of the authoritarian personality; particularly important is that R W A is defined
as a set of attitudes rather than as a personality type. Altemeyer defined R W A
in tenus of three clusters of attitudes that are similar to three of the characteristics
Adorno and his colleagues (1950) used to describe the authoritarian personality:
“authoritarian submission-a high degree of submission to the authorities who
are perceived to be established and legitimate in the society in which one lives;
236 CHAPTER 7
authoritarian aggression-a general aggressiveness, directed against various persons,
that is perceived to be sanctioned by established authorities; conventionalism-a
high degree of adherence to the social conventions that are perceived to he
endorsed by society and its established authorities” (Altemeyer, 1994, p. 133; see
lable 7.1 for sample questionnaire items used to assess RWA). If one thinks of
prejudice as a form of nonphysical, symbolic aggression, these attitudes lead people
high in R W A to be prejudiced against groups that authority figures condemn and
that are perceived to violate traditional values,
[
People high in R W A tend to be prejudiced against a wide variety of groups,
including feminists (Duncan, Peterson, & Winter, 1997), lesbians and gay men
(Whitley & Lee, 2000), Native Americans (Altemeyer, 1998), Muslims (Cohrs,
Moschner, Maes, & Kielman, 2005), immigrants (Quinton, Cowan, & Watson,
.}996), and fat people (Crandall, 1994). However, whereas some research shows
that people high in R W A are prejudiced against Mrican Americans (for example,
Altemeyer, 1998; Rowatt & Franklin, 2004), other research does not (for example,
Whitley, 1999). R WA has also been found to be related to prejudice not only in the
United States and Canada, but in other parts of the world as well, including Australia
and New Zealand (Duckitt, 2001; Heaven & St. Quintin, 2003), Western Europe
(Duriez & Van Hiel, 2002), Russia (McFarland, Ageyev, & Djintcharadze, 1996),
and South Africa (Duckitt, Wagner, du Plessis, & Birum, 2002).
Several personal characteristics of people high in R W A may predispose them
[to prejudice. First, people high in R WA tend to be mentally inflexible. They see
the world in simple terms, want definite answers to questions, and have a high
need for closure, especially when dealing with issues that are important to them
TAB L E 7.1 Sample Questionnaire Items Used to Assess Right-Wing
Authoritarianism (RWA)
Authoritarian Submission
It is always better to trust the judgments of the proper authorities in government and
religion than to listen to the noisy rabble-rousers in our society who are trying to create
doubt in people’s minds.
Once our government leaders give us the “go ahead,” it will be the duty of every patriotic
citizen to help stomp out the rot that is poisoning our country from within.
Authoritarian Aggression
What our country really needs is a strong, determined leader who will crush evil, and take
us back to the true path.
The situation in our country is getting so serious, the strongest methods would be justified
if they eliminated the troublemakers and get us back to our true path.
Conventionalism
The “old-fashioned way” and “old-fashioned values” still show the best way to live.
Our country needs free thinkers who will have the courage to defy traditional ways, even
if this upsets many people. a
aAgreement with this item indicates low RWA.
SOURCE: Altemeyer (1998, pp. 49–51).
INDIVIDUAL DIFFERENCES AND PREJUDICE 237
(Van Hiel, Pandelaere, & Duriez, 2004). As we saw in Chapter 4, this type of
mental inflexibility is associated with a propensity for stereotyping. Perhaps as a
reflection of this inflexibility, people high in R WA are uninterested in political
issues (peterson, Duncan, & Pang, 2002) and experiencing new things (Heaven &
Bucci, 2001), and so are unlikely to be exposed to views that differ from their own.
People high in R W A also tend to see the world as a dangerous and threatening-[
place, leading them to place a high value on security. They submit to authority_
and conform to group nonns as a way of finding security in the protection of the
group under the guidance of its authority figures (Duckitt, 2001).
In addition, people high in R WA tend to organize their worldviews in terrns~l
of ingroups and outgroups (Altemeyer, 1981, 1998). As we discuss in Chapter ~
strong identification with an ingroup promotes prejudice against outgroups, in part
by leading people to exaggerate the differences between the ingroup and outgroups.
These perceived differences can lead to the belief that outgroups threaten
the traditional values embraced by people high in R W A (see the discussion of
perceived value differences later in this chapter). By derogating outgroups, people
can dismiss them as unimportant and therefore as constituting no real threat to
ingroup values. Altemeyer (1981, 1998) also noted that people high in R WA
tend to be self~righteous, seeing themselves as more moral than other people and
therefore as justified in looking down on anyone authority figures define as less
moral than themselves. They may feel especially free to express prejudice against
members of outgroups, such as lesbians and gay men, who authority figures
condemn as immoral threats to traditional values.
An important aspect of R W A as a theory of prejudice is the role authority
figures play. People high in R W A accept as legitimate prejudice against groups
authority figures condemn, but not necessarily other fonns of prejudice. For exam~
pIe, some studies have found that people high in R W A have negative attitudes
toward lesbians and gay men but not toward African Americans (for example,
Whitley, 1999). This difference in anitudes is explainable in terrns of authority:
some religious and political authority figures condemn lesbians and gay men for
violating traditional values. However, most religious and political authority figures
do not condemn African Americans; instead, they actively oppose racial prejudice.
Because people high in R W A also tend to hold traditional religious beliefS (for
example, Spilka, Hood, Hunsberger, & Gorsuch, 2003), they are especially
responsive to the directions religious authorities set.
The importance of authority was demonstrated in a study conducted of uni~
versity students in what had been West Germany before reunification (petersen &
Dietz, 2000). The research consisted of a personnel selection simulation in which
the participants had to choose three candidates for a managerial position; half the
candidates were from the fonner West Gennany and half were from the fonner
East Germany, a group that was often the target of prejudice by fonner West
Germans. The participants were categorized as high or low in R W A and assigned
to one of two experimental conditions. In one condition a memo fronl the corn~
pany president indicated that he did not think that hiring a fomler East Geffilan
would be a good idea; in the other condition, the memo did not mention the
candidates’ regional background. Results showed that in making their selections,
238 CHAPTER7
participants low in R W A did not discriminate on the basis of regional background,
nor did participants high in R W A whose memo did not mention regional background.
However, participants high in R W A who thought the company president
did not want to hire a fanner East German recommended fewer former East
German candidates than did the other participants.
In summary, people high in R WA tend to be prejudiced against a wide variety
of groups, especially those that they perceive to violate traditional values and groups
that authority figures condemn. A number of psychological characteristics may predispose
people high in R WA to prejudice, including mental inflexibility, a disinterest
in experiencing new things, a perception of the world as a dangerous place, and a
tendency to organize their worldviews in tenus of ingroups and outgroups.
Social Dominance Orientation
Social dominance orientation (SDO) is an individual difference variable that
reflects “the extent to which one desires that one’s in-group dominate and be
superior to out-groups” (pratto, Sidanius, Stallworth, & Malle, 1994, p. 742). It
is com_prised of two closely related components, group-based dominance and opposition
to equality (lost & Thompson, 2000). Group-based dominance reflects
the belief that one’s group ought to be at the top of the societal ladder and that
other groups ought to be on the bottom; opposition to equality reflects the belief
that the groups on the bottom ought to stay there. People high in SDO believe
that the groups they identify with, such as racial or ethnic groups, socioeconomic
status groups, and so forth, should have a superior position in society and control
over society’s resources and that other groups should “stay in their place” and not
ask for more than they have. Thus, people high in SDO prefer a society in which
social groups are unequal and their group holds the superior position (Sidanius &
Pratto, 1999). See Table 7.2 for sample questionnaire items used to assess SDO.
[
– Not surprisingly, members of groups that hold more power in society exhibit
higher levels of social dominance orientation. For example, in the United
–States, Whites score higher than members of minority groups, men score higher
TAB L E 7.2 Sample Questionnaire Items Used to Assess Social
Dominance Orientation (sDO)
Group-Based Dominance (GBD)
It’s probably a good thing that certain groups are at the top and other groups are
at the bottom.
Inferior groups should stay in their place.
Opposition to Equality (OEQ)
We should do what we can to equalize conditions for different groups. a
Increased social equality would be a good thing.a
“Agreement with these items indicates low OEQ and low SDO.
NOTE: Items are from Sidanius and Pratto (1999, p. 67); classification of items as GBD and OEQ is from Jost and
Thompson (2000, p. 216).
INDIVIDUAL DIFFERENCES AND PREJUDICE 239
than women, heterosexuals score higher than lesbians and gay men, and the
wealthy score higher than the less wealthy; similar patterns have been found in
other countries (Pratto, Sidanius, & Levin, 2007; Sidanius & Pratto, 1999). In
addition, the longer people are members of a higher-power social group, the
higher they score on SDO. For example, Serge Guimond and his colleagues
(Guimond, Dambrum, Michinov, & Duarte, 2003) measured SDO in first-year
and upper-year students in a high-social-power profession-law-and in a lowsocial-
power profession-psychology. They found that law students’ SDO scores
increased with years in college whereas psychology students’ SDO scores decreased
with years in college. In addition, people high in SDO tend to be attracted to
high-power professions (Sidanius & Pratto, 1999). The link between social status
and SDO has been further demonstrated in experiments in which research participants
have been randomly assigned to high- or low-power roles. Participants
assigued to high-power roles score higher on SDO than do participants assigued to
low-power roles (Guimond et al., 2003). Therefore, social power is not simply
correlated with SDO; social power causes people to develop social dominance
attitudes. Thus, SDO is related to social power in two ways: People high in
SDO are attracted to high-power professions and socialization into the profession
increases SDO (Guimond et al., 2003).
Social Dominance Orientation and Prejudice. Given SDO’s roots in the de”:’-sire
to maintain social inequality, it is not surprising that people high in SDO are
prejudiced against members of groups that challenge the legitimacy of social inequality,
including racial or ethnic groups such as African Americans, Asian
Americans (Sidanius & Pratto, 1999), Native Americans (Altemeyer, 1998), Muslims
(Cohrs et al., 2005), and Australian Aborigines (Heaven & St. Quintin, 2003),
immigrants (Esses, Jackson, & Armstrong, 1998); lesbians and gay men (Whidey &
Lee, 2000); and feminists (Sidanius & Pratto, 1999). SDO has been found to be
related to prejudice not only in the Uuited States, but also in many other parts
of the world, including Australia and New Zealand (Ouckitt, 2001; Heaven &
St. Quintin, 2003), Western Europe (Ouriez & Van Hiel, 2002), Asia (Pratto
et aI., 2000), Israel (Levin & Sidanius, 1999), and South Africa (Ouckitt et al.,
2002). Thus, like R WA, SDO is related to multiple forms of prejudice in multiple
cultural contexts.
Like people high in R W A, those high in SDO have a number of personal
characteristics that may predispose them to prejudice. For example, people high
in SDO tend to see the world as what John Duckitt (2001) describes as a
“competitive jungle characterized by a ruthless and amoral Darwinian struggle
for survival, … in which might is right, and winning is everything” (p. 51). In
addition, people high in SDO tend to see resources as being in limited supply, so
that if someone else gets something, they lose out on it; they have trouble
believing that there could be enough for everyone (Esses et al., 1998). Taken
together, these characteristics motivate people high in SDO to try to deny resources
to nlembers of outgroups and to try to keep outgroups from gaining
any power that might force the sharing of resources. People high in SDO are
also tough-minded (Ouckitt, 2001) and low in empathy (McFarland, 2001); as
240 CHAPTER 7
we saw in Chapter 5, being able to empathize with members of other groups
tends to reduce prejudice.
[
– An important aspect of the social dominance theory is the concept of legiti~
mizing myths. Legitimizing myths are sets of attitudes and beliefs that people
high in SDO can use to justify their dominant position in society (Sidanius &
-Pratto, 1999). In the context of prejudice, group stereotypes are legitimizing
myths that can be used to justiry denying equality to other groups despite the
fact that prejudice is socially disapproved. For exam_pIe, the beliefs that members
of another group are lazy and of low intelligence could be used to justifY denying
equal educational opportunity and powerful positions in society to Inembers
of the stereotyped group: The logic of social dominance asks, why should society
expend precious resources to provide people with opportunities they are inherently
unfit to take advantage of? Consequently, being high in SDO leads people
to endorse stereotypes of outgroups, especially negative stereotypes, and these
negative beliefs then lead to prejudice. For example, Bernard Whitley (1999)
found that SDO was positively correlated with endorsement of both positive
and negative stereotypes of African Americans. How would positive stereotypes
contribute to the goal of keeping other groups down? Recall from Chapter 6 that
pos.itive stereotypes can contribute to that goal if they place people in low power
roles, such as by stereotyping African Americans as athletes and entertainers rather
than as business executives or government leaders. Whitley also found that when
endorsement of stereotypes of Mrican Americans and oflesbians and gay men was
controlled, the relationship between SDO and other indicators of prejudice was
greatly reduced. That is, among people high in SDO, those who endorse legitimizing
myths to a greater degree are more prejudiced. These results suggest that
legitimizing lllyths, in the form of stereotypes, are necessary for people high in
SDO to justiry their other prejudiced responses.
Recall from Chapter 4 that Stephanie Goodwin and her colleagues
(Goodwin, Gubin, Fiske, & Y zerbyt, 2000) found that people high in social power
tend to use stereotypes to a greater extent than do people low in social power.
They atnibuted this greater stereotype use to people high in power not being motivated
to individuate others. In a similar vein, Guimond and his colleagues (2003)
have found that power correlates with prejudice: People high in power express
more prejudice against a variety of outgroups. Guimond and colleagues explain
this finding in tern1S of SDO: People higher in social power are higher in SDO,
which leads to higher levels of prejudice. Social power, then, may potentiate
both SDO and stereotyping, with people high in SDO using those stereotypes
to justifY their prejudices. SDO may also have motivational effects; see, for
example, Box 7.1.
Social Dominance Orientation and Authoritarianism. In some ways SDO
and authoritariamslll sound very similar, both being ideological variables that
{-predispose people to prejudice, but they are, in fact, different. The most funda\\
lllental difference is that authoritarianism focuses on submission to ingroup
authority figures regardless of whether they advocate dominance over other
I groups, whereas SDO focuses on dominance over outgroups regardless of
\
The relationship between social dominance orientation
(500) and prejudice can take a number of forms.
Although we have focused on SOO as a potential cause
of prejudice, Henry Danso and Victoria Esses (2001)
took a different perspective, viewing 500 as a motive
that can be aroused given the right circumstances.
They reasoned that if SOO is based on a need to
maintain dominance over other groups, people high
in SOO should be motivated to prove their group’s
dominance, even if they are unaware of that
motivation.
In their study, Oanso and Esses had either a Black
or White research assistant individually administer a
standardized test of arithmetic ability to White college
students. The researchers reasoned that students high
in SOO would be motivated to show that Whites are
intellectually superior to Blacks and so would do better
on the test when it was administered by a Black
research assistant; students tested by a White assistant
and low SOO students tested by a Black assistant
should not differ from one another. Oanso and Esses
INDIVIDUAL DIFFERENCES AND PREJUDICE 241
found that the high SOO students who were tested by
a Black research assistant had an average score of
about 80 percent on the test, whereas the other
groups averaged only about 50 percent. Although SOO
is correlated with prejudice, Danso and Esses reported
that in a previous study there was no relationship
between level of racial prejudice and performance
when tested by a Black or White research assistant.
Therefore, the motivational effects found in their
study occurred as a result of 500, not prejudice.
The authors concluded that their findings “may
have practical implications for relations between groups
for whom there has previously been an unequal
distribution of power and resources (e.g., between men
and women in managerial positions or between Blacks
and Whites in the United States). In such situations,
perceived shifts in power balance may represent a
threat to the dominance of one group and, as a result,
motivate the dominant group members to work to
maintain their group dominance, especially if they de~
sire an unequal distribution of resources” (pp. 163-164).
the views of ingroup authority figures. That is, authoritarianism focuses onjJ relations vvithin groups (submission to ingroup authority) whereas SDO focuses
on relations between groups (dominance of the ingroup over outgroups). The
relationship between SDO and prejudice is higher for people who identifY more
strongly with their groups, supporting the intergroup nature of SDO (Wilson &
Liu, 2003), Stronger group identity motivates people to make stronger distinctions
between their group and other groups, to stereotype members of other groups, and
to view other groups less positively than one’s own group (see Chapter 9). The
importance of ingroup authority in authoritarianism was shown in Petersen and
Dietz’s (2000) study, described earlier, in which participants high in R WA acted in
accordance with an authority figure’s hint to discriminate against members of
an outgroup whereas participants low in R W A did not.
This difference in the nature of the two constructs is reflected in the low correlations
that are often found between scores on measures of SDO and R W A. For
example, Michele Roccato and Luca Ricalfi (2005) reported that the average correlation
between SDO and R W A was only r = .20 for studies conducted in the
United States and Canada, However, they also found that the average correlation
was much higher for people in Europe, Australia, and New Zealand. Theyattributed
the geographic differences in correlations to differences in political systems.
The countries with the higher correlations tend to make stronger distinctions between
the policies of the political left and those of the political right (see also
Duckitt, 2001). However, the correlation between SDO and authoritarianism is
not very high in any group.
242 CHAPTER 7
There are also differences in how SDO and authoritarianism relate to different
forms of prejudice. For example, SDO is linked to both racial and anngay prejudice,
whereas R W A is linked to antigay prejudice but not to racial prejudice
(Whitley, 1999). In addition, SDO is linked to hostile sexism but not to benevolent
sexism, whereas R W A is related to benevolent sexism but not to hostile
sexism (Sibley, Wilson, & Duckitt, 2007). This pattern reflects differences in the
nature of authoritarianism and SDO (Duckitt, 2001). Authoritarianism focuses on
perceived threats and obedience to authority figures’ nIles as a means of avoiding
those threats. Because lesbians and gay men are portrayed by some authority figures
as threats to important social values whereas African Americans are not, people
high in R W A respond to the purported threat with negative attitudes toward
lesbians and gay men but not toward African Americam. In contrast, benevolent
sexism represents endorsement of traditional gender roles and values, resulting in a
positive correlation benveen it and R WA. Turning to SDO, recall that it has two
components, opposition to equality and group based dominance. Because lesbians
and gay nlen are distributed across the socioeconomic spectrum, they do not present
a challenge to inequality in the distribution of society’s resources-they already
have economic parity with heterosexuals. African Americans, in contrast, do challenge
inequality. Hence, the opposition to inequality component of SDO is not
related to negative attitudes toward lesbians and gay men, but it is related to negative
attitudes toward African Americans (Whitley, 1999). Group-based dominance
is related to negative attitudes toward both lesbians and gay men and African
An1ericans because many heterosexuals classify lesbians and gay men as an outgroup
and many White Americans classify African Americans as an outgroup. Hostile sexism
portrays women, a traditionally subordinated group, as competing with men
for social status, thus evoking both the opposition to equality and group-based
dominance aspects of SDO.
In conclusion, then, authoritarianism and SDO represent two separate ideologically
based roots of prejudice (Duckitt, 2001). Authoritarianism focuses on seeking
security against perceived threats from other groups by confonnity to group norms
and obedience to authority. SDO focuses on quashing competition for resources
from other groups and maintaining the ingroup’s dominance in society.
SELF-ESTEEM
The self represents our awareness of ourselves as living beings who interact with
the world and the people in it. This awareness includes our beliefs about what
we are like, our characteristic behaviors, our abilities and shortcomings, and so
forth. Self-esteem refers to people’s evaluations of their personal characteristics and
behavioral patterns. People who evaluate themselves positively are said to have
high or positive self-esteem; people who evaluate themselves negatively are said
to have low or negative self-esteem. Researchers and theorists have proposed
two ways in which self-esteem might be related to prejudice, both of which can
operate at the same time (Crocker, Blaine, & Luhtanen, 1993). One role proposed
INDIVIDUAL DIFFERENCES AND PREJUDICE 243
for self-esteem is self-enhancement: looking down on others might make on~J
feel better about onesel£ The other role is self-protection: if one’s self-esteem
is threatened, looking down on others might again make one feel better about
oneself, especially if doing so can directly counteract the threat. For example, if
threat conles in the fonn of criticism from a member of a negatively stereotyped
group, viewing that group and its members as incompetent to make a valid criticism
can blunt its effect on self-esteem: If criticism comes from someone who is
incompetent to judge, it is meaningless and so is no reflection on oneself.
Self-Enhancement
Although the self-enhancement role of self-esteem seems to be quite straightforward
-one bolsters one’s self-image by looking down on others-research on the process
is complicated by the fact that the self-enhancement hypothesis can be interpreted in. ..
two ways (Aberson, Healy, & Romero, 2000). The traditional interpretation, based on general theories of self-esteem, is that people with low self-esteem should be moreJ
prejudiced than people with high self-esteem. It holds that “low self-esteem individuals
need to make up for poor self-concept, and therefore they may pick on others to
raise deficient self-esteem, whereas high self-esteem individuals do not need to bolster
self-esteem” (Aberson et al., 2000, p. 158). This interpretation postulates a negative
correlation between self-esteem and prejudice: People with low self-esteem should
be more prejudiced than people with high self-esteem. The alternate interpretationJ
is just the opposite: People with high self-esteem should be more prejudiced than
people with low self-esteem because prejudice is one source of self-esteem. As
Christopher Aberson, Michael Healy, and Victoria Romero (2000) put it, “bias allows
high self-esteem individuals to create, bolster, and maintain positive … identities. Low
self-esteem individuals have low self-esteem because they do not regularly engage
in … bias” (p. 158).
Researchers have used two approaches to investigate the self-enhancement
hypothesis. In one approach, they create artificial groups in laboratory settings. As
we saw in Chapter 3, even arbitrarily assigning people to artificial groups elicits
group loyalty that leads them to see their own group in more favorable tenus
than other groups. Researchers then can use this method to look at the degree to
which group members’ personal self-esteem is correlated with their ingroup bias.
The second approach uses survey research to assess the correlation between people’s
level of self-esteem and prejudice against minority groups in society. Aberson
and his colleagues (2000) reviewed the research using the first approach and found
an average correlation of r = .20 between self-esteem and intergroup bias. That is,
people with high self-esteem showed more bias than people with low self-esteem,
but only to a small degree. Survey research has produced inconsistent results:
Some studies have found small positive correlations between self-esteem and prejudice
(for example, Utsey, McCarthy, Eubanks, & Adrian, 2002) but others have
found negative correlations (for example, Little, Murry, & Wimbusch, 1998;
Valentine, 1998).
What causes these contradictory results? There has been too litde research to
know for certain, but there are several possibilities. One is that both high- andl
244 CHAPTER 7
low-self-esteem people are prejudiced, but that they express their prejudices in
clifferent ways (Aberson et al., 2000; Crocker et al., 1993). For example, highself-
esteem people may express their prejudice directly, such as by saying negative
things about outgroups. Their high self-esteem buffers them against any criticism
they may receive from expressing prejudice directly. In contrast, low-self-esteem
people already have low opinions of themselves and so want to avoid such criticism.
They therefore express their prejudice indirectly, such as by giving undeserved
low ratings to products created by members of outgroups or by avoiding
interaction with them. Because most research has used direct measures of bias,
researchers may have overlooked these indirect indicators of prejudice and so
r-may have erroneously concluded that high self-esteem people are more biased. A second possibility is that self-esteem is related to prejudice, but that the re\
lationship is indirect rather than direct. That is, self-esteelll might work through
some other variable to influence prejudice. For example, Jane Simoni (1996)
found only a small correlation between self-esteem and attitudes toward lesbians
and gay men, but larger correlations between both self-esteem and contact with
lesbians and gay nlen and between contact and attitudes toward lesbians and gay
men. She showed that higher self-esteem could lead to more contact with lesbians
and gay men and that more contact could lead to more positive attitudes
(see Chapter 14 for more discussion of intergroup contact and prejudice). Thus,
low self-esteem may make people reluctant to engage in the intergroup contact
that could lead to more favorable attitudes.
\’ A final possibility is that there are two kinds of high self-esteem, with one
L?eing related to prejudice and the other not. Christian Jordan, Steven Spencer,
and Mark Zanna (2005) have distinguished between what they call secure high
self-esteem and defensive high self-esteem. People with secure high self-esteem
truly have positive opinions of themselves whereas people with defensive high
self-esteem act as though they see themselves positively as a way to hide the
fact that they really doubt their self-worth. However, both types of people get
high scores on traditional measures of self-esteem. Using a measurement strategy
that distinguished between the two types of self-esteem, Jordan and his colleagues
found that people with defensive high self-esteem expressed more intergroup
bias than people with secure high self-esteem. These findings are quite
consistent with the self-enhancement role of self-esteem in prejudice: People
who are secure in their high self-esteem have no need to be biased, but those
who doubt their self-worth use bias as a means of bolstering their self-inlages.
Self-Protection
[
—“If prejudice is used to protect self-esteem, then a threat to self-esteem should lead
-” _!o increases in prejudice. Exhibiting prejudice would reduce the effects of the I L threat and return self-esteem to its prethreat level. Unlike the results of research on L self-enhancement, those for self-protection have been strongly supportive. We saw
some examples of the self-protective role of prejudice in Chapter 4, in which we
discussed research on how threats to self-esteem affected stereotype activation and
application (Fein & Spencer, 1997; Spencer, Fein, Wolfe, Fong, & Dunn, 1998).
INDIVIDUAL DIFFERENCES AND PREJUDICE 245
Steven Fein and Steven Spencer (1997) also demonstrated the role of prejudice
in self-esteem maintenance. Participants in their study took a bogus intelligence
test, after which some were told that they had done poorly (the self-esteem threat
condition) and others were told that they had done well (the no threat condition).
All participants then completed a self-esteem measure. Then, in what was ostensibly
another experiment, they evaluated either an Italian or Jewish job candidate
(pretesting had shown that there was a fair amount of anti-Jewish prejudice on
the campus where the research was conducted, but not much anti-Italian prejudice).
The participants in the self-esteem threat condition made more negative
ratings of the Jewish candidate than of the Italian candidate; participants in the no
threat condition rated both candidates equally highly. Similarly, Steven Fein and
colleagues (Fein, Hoshino-Browne, Davies, & Spencer, 2003) found that male research
participants whose self-esteem had been threatened sat farther away from a
man they thought was gay than a man they thought was straight; there was no
difference in seating distance for participants in a no threat condition. Note that
in both of these studies the threat to self-esteem came from the researcher, not
from a Jewish person in the first study or a gay man in the second; thus, a threat
from any source, not just the target of prejudice, can arouse prejudice.
The results of Fein and Spencer’s (1997) research also demonstrated the buffering
role of prejudice. They found that participants who rated the Jewish job
candidate and whose self-esteem had been threatened showed an increase in
self-esteem after making their ratings whereas the self-esteem of the other participants
did not change. Putting all their data together, then, Fein and Spencer
first showed that the threat to self-esteem caused lower ratings of the Jewish candidate
and then showed that those lower ratings were associated with increased
self-esteem. That is, expressing prejudice warded off the threat to self-esteem.
Examining this issue from a different perspective, Fein and his colleagues (2003)
reinforced the self-esteem of some research participants before giving them an
opportunity to express their opinions about students attending their college and
students attending a rival college. The participants whose self-esteem was not
reinforced showed the typical ingroup bias, rating students at their college as better
than students at the rival college; however, the students whose self-esteem
had been reinforced showed no such bias. Thus, reinforcing self-esteem seems
to inoculate participants against normal intergroup bias and prevents prejudice.
Finally, it appears that prejudice can have self-protective effects in children
as well as adults. Jayne Stake (2003) studied high school students attending a
science enrichment program. Focusing on male students, she assessed their selfconfidence
in their science abilities and their attitudes toward women in science
at the beginning of the program and at its end four weeks later. Students were
exposed to positive information about women in science and to women scientists
as role models throughout the program. At the beginning of the program,
Stake found that boys with lower science self-confidence tended to have
negative attitudes toward women in science. However, boys whose science
self-confidence increased over the course of the program had more positive attitudes
toward women in science at the end of the program than at the beginning.
Apparently, the initial negative attitudes toward women in science functioned to
246 CHAPTER 7
protect the boys’ self-esteem as science students. As that self-esteem increased
over time, the protection that prejudice provided was no longer needed and
their attitudes became more positive.
PERSONAL VALUES
Values are the enduring beliefs people hold concerning the relative importance of
the goals they aspire to achieve in life and the types of outcomes they should try
to avoid (Rokeach, 1973). For example, people who place a higher value on security
than on freedom will do all they can to protect themselves and their families
from harm, even if it means giving up some freedom. In contrast, people who
place a higher value on freedom than on security will do all they can to ensure
that others have minimal control over them, even if it means taking risks and
therefore giving up some security. In addition to defining goals, values also serve
as standards for making evaluative judgments: People and things one perceives to
be consistent with one’s values are judged to be good and those perceived to be
inconsistent with one’s values are judged to be had (Schwartz, 1996).
Psychologists have related values to prejudice in several ways. Some theories,
such as Irwin Katz and Glen Hass’s (1988) theory of ambivalent prejudice,
hold that values are directly related to prejudice: Some values facilitate prejudice
whereas other values inhibit prejudice. Other theories, such as Milton Rokeach’s
(1972) value dissimilarity model and the theory of modem-symholic prejudice
(McConahay, 1986; Sears & Herny, 2005), hold that prejudice arises because people
believe that outgroup members hold values that are incompatible with or threaten
those of the ingroup. Finally, the theory of aversive prejudice (Dovidio & Gaertner,
2004) holds that egalitarian values, or beliefs in equality, lead White people to
suppress overt prejudice against other groups and to redirect it into more subde
fOnTIs, such as avoidance of intergroup contact (see Chapter 6 for discussion of these
fonus of prejudice). In this section, we first examine the direct relation of values to
prejudice and then look at some theories based on perceived value dissimilarity.
Value Orientations
Several theories postulate that two general categories of values are related to prejudice,
although different theories give different names to the values (Sampson,
1999). One category, generally referred to as individualism, relates to values emphasizing
the importance of self-reliance; the other category, generally referred to as
egalitarianism, emphasizes the importance of all people being treated equally and
fairly (Katz & Hass, 1988).
l Individualism. Historically, individualism is a value that has long been important
in North America (Kinder & Mendelberg, 2000). North Americans place a
_ strong emphasis on self-reliance and independence from others (Biernat, Vescio,
-Theno, & Crandall, 1996). Donald Kinder and Tali Mendelberg (2000) explain
INDIVIDUAL DIFFERENCES AND PREJUDICE 247
that during the 19th century, individualism came to be associated with hard work
as the route to success in life because the wealth obtained through hard work
allowed one to be independent of others and to do whatever one chose. At the
same time, idleness came to be seen as a vice. As a result, “in America today, idleness
is … a moral defect; hard work, in and of itself, a moral virtue; dependence
on others, a disreputable condition” (Kinder & Mendelberg, 2000, p. 47). Because
of this link between individualism and hard work, most research on individualism
defines the concept in terms of what is called the Protestant ethic or Protestant work
ethic, which emphasizes the importance of hard work and perseverance as the way
to success in life (Fumham, 1990). Although there are many measures of the
Protestant ethic, the scale devised by Katz and Hass (1988) is one of the most
commonly used; the first section of Table 7.3 contains some sample items from
the scale.
Group stereotypes provide the link between individualism and prejudice”J’
Groups that are stereotyped as behaving in ways that violate the principles of
individualism are viewed negatively by those who adhere to these principles
(Biernat et aI., 1996). Thus, Monica Biernat and her colleagues (1996) found
that people who score high on the Protestant ethic hold negative attitudes toward
African Americans and fat people, two groups that are stereotyped as lazy,
although the correlation is stronger for African Americans as the target group
(see Katz & Hass, 1988, and Sears & Henry, 2005, for other examples of racial
attitudes and Crandall, 1994, for anti-fat attitudes). Interestingly, Biernat and her
colleagues also found a correlation between work ethic scores and negative attitudes
toward gay men. This correlation was of about the same magnitude as
that for attitudes toward African Americans even though gay men are not stereotyped
as lazy. This may be because the Protestant ethic includes values such as
self-restraint and avoidance of pleasure seeking and gay men are often stereotyped
TAB L E 7.3 Sample Questionnaire Items Used to Assess Values
Individualism/Protestant Ethic
Respondents rate the extent to which they agree or disagree with each item:
Most people who don’t succeed in life are just plain lazy.
Anyone who is willing and able to work hard has a good chance of succeeding.
If people work hard enough they are likely to make a good life for themselves.
A distaste for hard work usually reflects a weakness of character.
Egalitarianism
Respondents rate the extent to which they agree or disagree with each item:
There should be equality for everyone-because we are all human beings.
Those who are unable to provide for their basic needs should be helped by others.
Everyone should have an equal chance and an equal say in most things.
Prosperous nations have a moral obligation to share some of their wealth with
poor nations.
SOURCE: Katz and Hass (1988, p. 905).
248 CHAPTER 7
as hedonistic (Biernat et aI., 1996), Not all research supports a link between individualism
and prejudice. For example, Margo Monteith and Gina Walters (1998)
found essentially no correlation between endorsement of individualism and antiBlack
prejudice in a sample of White college students, and Gloria Cowan, Livier
Martinez, and Stephanie Mendiola (1997) found that individualism did not correlate
with non-Latino college students’ attitudes toward illegal Latino immigrants.
Thus, the actual link benveen individualism and prejudice may not be as strong as
some theories have proposed.
[
Egalitarianism. As a value position, egalitarianism reflects a strong emphasis
on the principles of equal opportunity, equal treatment for all people, and concern
for others’ well-being (Biernat et al., 1996). The second section of Table 7.3 con~
tains some sample items from the scale most commonly used to assess endorsement
r
-of egalitarian values (Katz & Hass, 1988). In contrast to individualism, which is
held to facilitate prejudice, theorists propose that egalitarianism inhibits prejudice.
–As Biernat and her colleagues (1996) expressed it, White An’lericans who endorse
egalitarian values “either experience feelings of sympathy for Black Americans [as
proposed by the theory of ambivalent prejudice] or they work to avoid the threat
to self-concept that negative behavior toward Blacks would produce [as proposed
by the theory of aversive prejudice]. In either case, egalitarian values work as
brakes on racist reactions” (p. 154). In addition, whereas individualism is proposed
to affect prejudice only when the group stereotype includes characteristics that are
contrary to individualistic values, theorists propose that egalitarianism works to
counteract all fonus of prejudice: “It represents a fonn of antiprejudice that is not
specific to any particular group or underlying cause of negative affect toward
outgroups; it is a ‘prejudice antidote'” (Biernat et al., 1996, p. 155).
What, then, is the relationship betvveen egalitarianism and prejudice? Biernat
and her colleagues (1996) included egalitarianism as well as individualism in their
study of attitudes toward Mrican funericans, lesbians and gay men, and fat people.
Endorsement of egalitarian values was negatively correlated with prejudice against
each group; that is, greater endorselnent of egalitarian values was associated with
less prejudice (for other examples, see Cowan et al., 1997; Katz & Hass, 1988;
Monteith & Walters, 1998). Biernat and her colleagues also found that egalitarianism
was more strongly related to prejudice than was individualism. In addition,
using a different measure of values, LiJach Sagiv and Shalom Schwartz (1995)
found that endorsement of egalitarian values was positively correlated with Israeli
Jews’ willingness to interact with Israeli Arabs. Thus, as Biernat and her colleagues
(1996) proposed, egalitarianism does appear to be a general antidote to prejudice.
How does egalitarianism have its effects? Recall from Chapter 4 that stereotypes,
which in their negative form constitute one aspect of prejudice, must be
activated before they can have an effect on people’s thoughts about and behavior
toward outgroups. For examples, Gordon Moskowitz, Amanda Salomon, and
Constance Taylor (2000) have found that srimuli associated with outgroups are
less likely to acrivate stereotypes for people who strongly endorse egalitarian values
than for people who are less egalitarian. Thus, egalitarianism may inhibit prejudice
by preventing the activation of negative stereotypes.
INDIVIDUAL DIFFERENCES AND PREJUDICE 249
Perceived Value Differences
Rokeach (1972) proposed that prejudice is, in part, based on the perception that
outgroups’ value systems differ from one’s own. Because values guide judgments
of what is good or bad, holding different values implies a lack of goodness in the
outgroups. In a sense, this value difference hypothesis represents the mirror in1~-~
age of the well-established psychological principle that people like others who are
similar to them, especially if they are similar in tenus of abstract characteristics such as
attitudes and value positions (Berscheid & Reis, 1998). This principle also applies to
members of outgroups. For example, several studies have found that people who
learn that a gay man holds attitudes similar to their own like him better than people
who learn that his attitudes differ from theirs. This finding holds even for people
high in prejudice against gay men (for example, Pilkington & Lydon, 1997). Bear
in nlind, however, that liking one person more than another does not necessarily
mean liking that person a lot. These studies also found that expressed liking for a
similar gay man was at about the level of that expressed for a dissimilar heterosexual
man, and both were liked less than a similar heterosexual man. In this section, we
first look at research on the relation of perceived value differences to prejudice. We
then present two theoretical explanations for the relation of value differences to
prejudice-terror management theory and the attribution-value model.
Value Dissimilarity. The value dissimilarity hypothesis holds that one source of
prejudice is the belief that members of outgroups do not share the values of one’s
ingroup. In fact, most people assume that outgroup members’ values differ fronl
their own. For example, White Arnericans believe that Black AInericans are less
likely to share their important values and less likely to live their lives according to
those values than are other White Americans (Biernat et al., 1996); heterosexuals hold
similar beliefS aboutlesbians and gay men (Biernat et al., 1996). Perceptions of grouPJ~
differences in values have also been called symbolic beliefs (Haddock, Zarma, &
Esses, 1993) and symbolic threats (Stephan & Stephan, 2000). However, the last 1
term may be misleading in some cases; although Whites and heterosexuals believe
that Blacks and lesbians and gay men do not share their values, they do not always
believe that members of those groups violate or disrespect their values (Biernat et al.,
1996).
Nonetheless, a perception of a lack of common values seems to be sufficient forJ~~
prejudice. Perceptions of value differences are correlated with majority group prejudice
against minority groups in a number of contexts, including prejudice against
African Americans (Stephan et al., 2002); Cuban, Mexican, and Asian immigrants to
the Urtited States (Stephan et al., 1999); lesbians and gay men (Biernat et aI., 1996);
Native Canadians (Corenblum & Stephan, 2001); French and Pakistani Canadians
(Esses, Haddock, & Zarma, 1993); Russian immigrants to Israel (Bizman & Yinon,
2001); and fat people (Biernat et aI., 1996). Perception of vaIue differences is a twoway
street: It should be related to minority group members’ attitudes toward the
majority group as well as for majority group attitudes toward minority groups.
This pattern has been found for ratings of White Americans by Mrican Americans
(Stephan et a!., 2002), of White Canadians by Native Canadians (Corenblum &
Stephan, 2001), and of men by women (C. W. Stephan et aI., 2000).
250 CHAPTER 7
Most of the research cited above has measured value differences in general
terms, examining the extent to which outgroup attitudes are perceived to be different
from one’s own, The value dissimilarity hypothesis also holds that because
specific groups are sometimes stereotyped as violating specific values, prejudice
against those groups should be higher among people who endorse those values,
Supporting this idea, Biernat and her colleagues (1996) found that the extent to
which people gave beauty a high raring as a value was correlated with prejudice
against fat people. Perhaps the best known example of a values-prejudice relationship
is that of using family values as a justification for discrimination against
lesbians and gay men; see Box 7.2.
Since the 1992 presidential election, the concept of
family values has played a major role in political debates
in the United States, especially in debates about the civil
rights of lesbians and gay men (for example, Cloud,
1998), Although the concept of family values is poorly
defined (Cloud, 1998), lesbians and gay men are
stereotyped as violating those values, being perceived
as incapable of maintaining stable relationships, being
bad parents, corrupting children, and violating tradi~
tiona I gender roles that some people view as funda~
mental to family life (Vescio & Biernat, 2003). Because of
this perceived conflict between the gay/lesbian stereo~
type and the stereotype of the traditional family (for
example, McLeod & Crawford, 1988), one would expect
that endorsement of the traditional family would be
related to attitudes toward lesbians and gay men.
Surprisingly little research has been conducted on
this topic, but what research there is supports the
hypothesis. For example, Gregory Herek (1988) found
negative attitudes toward lesbians and gay men to be
correlated with scores on a measure of traditional
family ideology that focused primarily on parents as
the authority in the family and traditional husband~
wife and other gender roles. More recently, Theresa
Vescio and Monica Biernat (2003) examined college
students’ evaluations of a gay or heterosexual man
who was portrayed as either a good father or a bad
father. Participants who rated family security as an
important value evaluated the heterosexual father more
favorably than the gay father; participants who rated the
value as less important evaluated the two fathers
equally. Interestingly, whether the men’s parenting
behavior was consistent or inconsistent with traditional
family values had little effect on the ratings. That is, even
when the gay father’s behavior demonstrated support
for one aspect of traditional family values, effective
parenting, participants who said they valued the family
highly gave him a lower rating than a heterosexual
father who behaved in the same way.
Family values are also associated with attitudes
toward homosexuality in some non~Western cultures.
For example, Ming~Hui Hsu and Judith Waters (2001)
assessed the relationship between filial piety and
attitudes toward lesbians and gay men among Chinese
college students. Filial piety refers to “the highest
virtue within Confucian doctrine, … the production of
male offspring to maintain the family name [and] offer
sacrifices after death” (Hsu & Waters, 2001, p. 3). Hsu
and Waters found that greater endorsement of filial
piety was associated with more negative attitudes
toward both lesbians and gay men for both male and
female students. Thus, as the value dissimilarity model
would predict, people who strongly endorse a variety
of beliefs that can be categorized as family values hold
negative attitudes toward lesbians and gay men, a
group stereotyped as violating those values.
Finally, Dana Cloud (1998) reports an interesting
sidelight on the family values issue. She notes that its
first political use was in the context of race, not sexual
orientation. It occurred when, in May 1992, then Vice
President Dan Quayle said that racial unrest then
taking place “is directly related to the breakdown of
family structure” (quoted in Cloud, 1998, p. 395). In
fact, Cloud found that in political discourse from
December 1992 to July 1996, family values were
mentioned almost three times more often in a racial
context than in a sexual orientation context. Despite
this difference in use, there appears to be no research
on the relationship between endorsement of family
values and racial attitudes.
INDIVIDUAL DIFFERENCES AND PREJUDICE 251
Terror Management Theory. Jeff Greenberg, Sheldon Solomon, and Tom
Pyszczynski (1997; Pyszczynski, Solomon, & Greenberg, 2003) developed terror
management theory to explain (among other issues) how people’s desire to promote
and defend their belief and value systems results in prejudice. Their explanation
is rooted in two human characteristics-the instinct for self-preservation and
the contrasting knowledge that one’s death is inevitable. Greenberg and his colleagues
propose that the coexistence of the self-preservation instinct and the knowledge
of one’s vulnerability to death leads to terror because the self-preservation
instinct motivates people to try to avoid the unavoidable, death. As a species,
one way in which humanity has dealt with this terror is by developing cultural
institutions and worldviews that promise immortality. The pro:m:ised inunortality
can take two fonns. It can be literal, in the form of religious beliefs in an immorral
soul that lives on after physical death. Immortality can also be symbolic, in the
fonn of identification with time- and death-transcending social institutions such
as the £urrily and the nation and of tangible reminders of continuity such as children
and culturally valued achievements that carry on one’s reputation after death.
Because culture and its values provide a buffer against the terror created bY~death,
people are motivated to defend their culture against perceived challenges
to its validity, such as those posed by different cultural worldviews. If such challenges
were to succeed, they would undennine the protective cultural worldview
and leave people open to the terror created by the knowledge of death. The theOry]’
therefore proposes that if people are made aware of the inevitability of their own
deaths they will experience a need to reinforce their faith in their culture. One
fonn this reinforcement takes is the rejection of people who challenge the culture’s
beliefs and values or who represent other cultures: “The mere existence of alternative
[worldviews) will be psychologically unsettling, because granting their validity
either explicitly or implicitly undennines absolute faith in one’s own worldview ….
The most common response is to simply derogate either the alternative worldview
or the people who hold that view. By dismissing other worldviews as inaccurate, or
the people who hold such views as ignorant savages who would share our perspectives
if they were sufficiently intelligent or properly educated, the threat to one’s
own point of view is minimized” (Greenberg et aI., 1997, p. 70).
Researchers test the effects of awareness of one’s future death with an experimental
manipulation that induces what is called mortality salience. In response to a
supposed projective personality test, participants in the mortality salience condition
write a brief paragraph about what they think will happen to them when
they die and the emotions they feel while thinking about their own deaths.
Participants in the control condition typically write about a negative experience
that does not imply death, such as dental pain. Although this manipulation may
sound somewhat minimal, there is a substantial body of research attesting to its
effecriveness (Greenberg et al., 1997). Mter participants write their paragraphs,
researchers administer other manipulations and measure the dependent variables,
such as by having participants evaluate a person who either does or does not
challenge their worldviews.
Most research on terror management theory has focused on responses to
people who directly challenge participants’ worldviews and cultural values, such
252 CHAPTER 7
as someone who has written an essay challenging some aspect of traditional
American values (Greenberg et al., 1997); much less research has focused on reactions
to ethnic groups or other targets of societal prejudice. In one study that
did so (Greenberg et al., 1990), research participants who identified themselves as
Christians underwent a mortality salience manipulation, after which they read
what they were told were self-descriptions written by two other students at their
university, one of whom was depicted as a Christian and the other as Jewish.
Participants in the mortality salience condition rated the Christian student more
positively than the Jewish student; the religion of the students being rated did
not affect the evaluations of the participants in the control condition, Taking a
different approach to the assessment of prejudice, Lori Nelson and her colleagues
(Nelson, Moore, Olivetti, & Scott, 1997) manipulated mortality salience and
then had research participants read the case of a man who had been seriously
injured when his car crashed. The victim was suing the car’s manufacturer, alleging
that a manufacturing defect caused the accident; the car manufacturer was
described as being either All1erican or as Japanese. Participants in the mortality
salience condition assigned more blame to the Japanese manufacturer than to
the American manufacturer; there was no difference in the control condition.
Finally, Jeff Schimel and his colleagues (1999) examined the effects of mortality
salience on stereotyping. They hypothesized that because outgroup stereotypes
are components of cultural worldviews, participants experiencing 1110rtality
salience would respond favorably to an outgroup member who acted consistently
with the group stereotype (because such behavior would be consistent with their
worldview) and would respond unfavorably to an outgroup member who acted
inconsistently with the group stereotype (because such behavior would contradict
with their worldview). After undergoing a mortality salience manipulation, White
research participants read one of three essays purportedly written by a Black
student about his summer activities. In the stereotype-consistent condition, the
student reported engaging in such activities as “splitting to L.A., serious hoop,
slammin’ night life, cnusin’ for honeys, clubbing, getting stupid, a few run-ins,
drinking forties” (Schimel et al., 1999, p. 914). In the stereotype-inconsistent
condition, the writer used fannal language and told about taking SUffilner
engineering classes, working for a software company, and reading two novels
about World War II. A stereotype-neutral essay told about the student’s traveling
to San Francisco for sightseeing, to Ohio to visit family, and to New Orleans for a
friend’s wedding; he could afford the trips because his mother worked for an airline.
As shown in Figure 7.1, in the mortality salience condition, liking for the
Black student decreased as his behavior became less stereotype consistent; the opposite
pattern was found in the control condition. Another way of looking at
these results is that for control participants, liking increased with apparent value
similarity, which is consistent with the belief similarity effect discussed earlier;
however, for mortality salient participants, liking decreased with apparent value
similarity because that similarity contradicted the participants’ worldviews.
Taken together, the results of these studies indicate that mortality salience
Jeads to increased prejudice. What psychological mechanism underlies this effect?
r From a terror management theory perspective, prejudice and stereotyping of

2
Mortality salience
condition
FIG U R E 7.1 Mortality Salience and Stereotyping
INDIVIDUAL DIFFERENCES AND PREJUDICE 253
• Stereotype-consistent behavior
Stereotype-neutral behavior
• Stereotype-inconsistent behavior
Control condition
Under mortality salience conditions, White research participants liked a Black student more as his behavior became
more stereotypical. Under control conditions, White research participants liked a Black student less as his behavior
became more stereotypical.
SOURCE: Adapted from Schimel et al. (1999, Table 3, p. 914).
outgroup members reinforces people’s cultural worldviews. It does so by emphasizing
the negative characteristics of the outgroup that is challenging their worldview.
The presence of these negative characteristics implies that any challenge
the group makes is defective and therefore no real threat to the worldview.
Mortality salience also increases people’s feelings of identification with their ingroup
(Castano, Yzerbyt, Paladino, & Sacchi, 2002); as we discuss in Chapter 9,
strong ingroup identification tends to lead to prejudice. This increased identification
with the ingroup can result in a kind of “My group, right or wrong!”
mentality, leading people to tolerate negative behaviors by ingroup members
they would otherwise condemn, See Box 7,3 for an example,
The Attribution-Value Model. The attribution-value model (Crandall et ai” j 2001) proposes that prejudice begins with the perception that members of
minority groups have characteristics that are contrary to majority group values.
Thus, fat people are seen as lazy and unable to exercise restraint in eating, and
lesbians and gay men are seen as violating family values. Coupled to that perception
is the belief that members of those groups are responsible for their undesirable
characteristics. Because people who are seen as responsible for their negative
254 CHAPTER 7
Two of the effects of mortality salience are a motivation
to defend one’s cultural worldview by derogating other
groups (Greenberg et al., 1997) and increased identjfi~
cation with the ingroup (Castano et aI., 2002). As two
studies have shown, the combination of these factors
can lead people to tolerate acts of racism that they
would otherwise condemn. Jeff Greenberg and his
colleagues (Greenberg, Schimel, Martens, Solomon, &
Pyszczynski, 2001) had White research participants
undergo a mortality salience manipulation after which
they read an employment discrimination case in which
the plaintiff alleged that he was repeatedly passed
over for promotion because of his race. In one version
of the case, the employee was Black and the manager
who had allegedly blocked his promotion was White;
in the other version, these roles were reversed.
Participants rated the extent to which they thought
the manager was guilty of discrimination. Participants
in the mortality salience condition rated the White
manager as less guilty of discrimination and the
Black manager as more guilty of discrimination
compared to participants in the control condition.
Thus, mortality salience functioned to lessen the
perceived guilt of an ingroup member who harmed
an outgroup member and to increase the perceived
guilt of an outgroup member who harmed an
ingroup member.
Joel Lieberman and his colleagues (Lieberman,
Arndt, Person ius, & Cook, 2001) examined a more
extreme situation, that of hate crimes. Following a
mortality salience manipulation, research participants
read a summary of a crime in which two young men
attacked a man who had just left what was described
as a “Jewish Pride rally,” a “Gay Pride rally,” or just
“a rally.” The first two versions of the crime were
clearly hate crimes because witnesses reported that the
attackers shouted aggressive anti-Jewish or antigay
insults as they beat their victim; in the control
condition, nonspecific insults were used. The victim had
to be hospitalized because of the injuries he received.
After reading the case, participants recommended the
amount of bail that one of the alleged perpetrators
should be required to post. Participants in the control
condition recommended higher bail amounts for the
hate crimes than for the other assault; in the mortality
salience condition, lower bail was recommended for
the alleged hate crime perpetrator. As in the job
discrimination study, mortality salience lessened the
perceived guilt of an ingroup member who harmed an
outgroup member.
characteristics arouse negative emotions in others (Weiner, 1995), prejudice
results when groups are perceived to be responsible for their negative stereo typic
characteristics. Thus, studies conducted in a number of countries have found that
dislike of fat people is correlated with the belief that fatness is a matter of choice:
If fat people would only choose not to eat so much, they would not be fat
(Crandall et aI., 2001), Similarly, researchers have found that less negative attitudes
toward homosexuality are held by people who believe that homosexuality
is a matter of biology rather than choice (for example, Jayaratne et aI., 2006) or,
nl0re generally, that homosexuality is not sOluething that is changeable or under a
person’s control (for example, Haslam & Levy, 2006).
The research just cited was correlational and so could not show that attributions
of responsibility cause prejudice. However, some experimental research has
been conducted to examine whether manipulating people’s perceptions of the
cause of a negative characteristic results in differing evaluations of people with
that characteristic. For example, William Dejong (1980) had research participants
give their impressions of a young woman who was portrayed, through photographs
and infonnation in a fact sheet, as overweight. The fact sheet told some
participants that the woman’s overweight was due to a medical condition; the
fact sheet read by other participants provided no cause for the overweight,
INDIVIDUAL DIFFERENCES AND PREJUDICE 255
implying that it was due to overeating. A third group of participants rated an
average weight woman. The woman who was overweight due to a medical condition
was rated more favorably than the overweight woman without a nledical
condition and equally favorably as the average-weight woman. Therefore, leading
people to believe that a negative characteristic was caused by something not
under the person’s controlled to more favorable ratings of the person. Similar
results have been found for ratings of a person whose offensive body odor was
attributed either to a medical condition or to not bathing (Levine & McBurney,
1977) and for attitudes toward lesbians and gay men after participants read about
homosexuality being either a matter of biology or an undetermined cause
(piskur & Degelman, 1992).
Although strong evidence shows that the attributions for the cause of obesity
and homosexuality are related to attitudes toward those groups, the attributionvalue
model might not apply equally well to all forms of prejudice. For example,
Nick Haslam, Louis Rothschild, and Donald Ernst (2002) found that beliefS about
whether a group’s characteristics are changeable were related to attitudes toward
gay men but not to racial or gender attitudes. This difference may exist because of
two dimensions people use to classifY social groups (Haslam et al., 2000). One
dimension is naturalness} the extent to which group membership is seen as biologically
based. If a group is seen as highly natural, group nlembers cannot leave their
group for another; for example, a woman cannot change into a man. However,
members of low-naturalness groups can change their membership; for example, a
Republican could become a Democrat. The other dimension is entitativity} the
extent to which group members are seen as being similar to each other, to which
knowing that a person is a member of the group provides useful infonnation
about the person, and to which group membership is exclusive (the person either
clearly belongs to the group or not; there is no in-between state). For example,
all members of a political party are assum_ed to hold the same political views, people
assume that knowing a person’s party affiliation provides information about
those views, and a person is either a Republican or Democrat, not both at the
same time.
Some groups, such as racial and gender groups, are seen as being high on both
dimensions. Femaleness, for example, is seen as both natural in a biological sense
and as entitative: all women are perceived to share certain characteristics, knowing
that a person is a woman rather than a man supposedly provides useful infonnation
about her, and (most people believe) a person must be either a man or a woman:
there is nothing in between. Other groups, such as lesbians and gay men, are seen
as high on entitativity but low on naturalness. That is, they are seen as coherent
social groups, but not as natural or biological in nature, and so people can change
from one group to another (for example, from homosexual to heterosexual).
Therefore, being gay or lesbian is seen as a matter of choice, just as one can choose
to be a Republican or Democrat (political groups are also seen as low on naturalness
but high on entitativity). Haslam and his colleagues (2002) therefore suggest
that attributional models of prejudice apply only to groups that are seen as high on
entitativity and low on naturalness: “Categories that are represented as unambiguous
natural kinds–……..such as races and genders-cannot be understood in tenus
256 CHAPTER 7
of personal control and choice. Their members cannot be held responsible for
belonging to them if membership is a matter of immutable biology …. By this
account, prejudice towards Uesbians and] gay men is more strongly associated
with [entitativity] than are sexism and racism because the culture’s prevailing belief
that homosexuality is not a natural kind allows a particular fann of stigmatizing”
based on attributions of responsibility for violating cultural values (pp. 9&-97).
SOCIAL IDEOLOGIES
Ideologies are sets of attitudes and beliefs that predispose people to view the
world in certain ways and to respond in ways consistent with those viewpoints.
As John Jost and colleagues (jost, Glaser, Kruglanski, & Sulloway, 2003) explain,
ideologies are psychologically important for two reasons. First, “people adopt
ideological belief systems … to satisfY their psychological needs and motives”
(p. 341). For example, authoritarianism appeals to people with strong needs for
structure and certainty; authority figures can provide both (for example, Altemeyer,
1996). Second, “people embrace ideological belief systems at least in part because
they inspire conviction and purpose” (p. 351); that is, they give people
goals to strive for. Two important social ideologies that have been studied
in relation to prejudice are religion and political orientation.
Religion
More than half a cennuy ago, Gordon Allport (1954) wrote that “the role of religion
[in prejudice] is paradoxical. It makes prejudice and it unmakes prejudice ….
The sublimity of religious ideals is offset by the horrors of persecution in the name
of these same ideals” (p. 444). The situation has not changed much since then: as
we will see, some fonus of religiosity (ways of being religious) are positively correlated
with prejudice and other fonus of religiosity are negatively correlated with
prejudice. We first examine the relationship of degree of religious involvelnent
to prejudice, continue with the concept of religious orientation, take a bit of a
detour to look at the concepts of proscribed and permitted prejudices, and conclude
with religious fundamentalism.
As you read this section, there are two cautions to bear in mind. First, alnloSt
all the research on the relationship between religion and prejudice has been conducted
in North America and most of the participants in the research have been
White, middle-class Christians (Batson, Schoenrade, & Ventis, 1993). Therefore,
little is known about the relationship of religious faith to prejudice among believers
in other religions such as Buddhism, Hinduism, Islam, and Judaism, to name
only a few of the world’s other large religions. Second, because most of the focus
has been on the degree or nature of research participants’ religiosity, nomeligious
people have rarely been included in the research (Batson & Buntis, 1994).
Consequently, we know relatively little about the ways in which nonreligious
people differ from religious people and how those differences relate to prejudice.
INDIVIDUAL DIFFERENCES AND PREJUDICE 257
Religious Involvement. Because almost all religions teach intergroup tolerance,
one would expect that people who are nlOre involved in their religions, and so
presumably believe their religion’s teachings more strongly, would show less prejudice.
However, based on a review of 38 studies conducted from 1940 to 1990,
Daniel Batson and his colleagues (1993) found that religious involvement was consistendy
correlated with a variety of fonns of prejudice. Research conducted since
1990, which has focused largely on attitudes toward homosexuality, has found a
similar pattern of results (Whitley, in press). Thus, it appears that, as Batson anU
his colleagues (1993) wrote, “religion is not associated with increased love and
acceptance but with increased intolerance, prejudice, and bigotry” (p. 302).
Intrinsic and Extrinsic Religious Orientation. These findings were not received
well by religious leaders and researchers interested in the psychology of religion,
who had expected that there would be a negative correlation between religious
involvement and prejudice (Batson et aI., 1993). As a result, the focus of research
shifted from the degree to which people are religious to the way in which they
are religious, that is, from quantity of religious involvement to quality of religious
involvement. This shift is consistent with a distinction Allport (1954) had
made some years earlier: “belonging to a church because it is a safe, powerful,
superior in-group is likely to be the mark of an authoritarian character and to be
linked with prejudice. Belonging to a church because its basic creed of brotherhood
expresses the ideals one sincerely believes in, is associated with tolerance”
(pp. 452-453). These ideas evolved into the concepts of intrinsic and extrinsic
religious orientation (Allport & Ross, 1967). .”
People with an intrinsic religious orientation truly believe in their reli-j
gions’ teachings and try to live their lives according to them. They “find their
master motive in religion …. Having embraced a creed the individual endeavors –
to internalize it and follow it fully. It is in this sense that he lives his religion’:_
(Allport & Ross, 1967, p. 434, emphasis in original). People with a strong internal 1
orientation should be unprejudiced to the extent that their religions teach inter- I
group tolerance, In contrast, people with an extrinsic religious orientation use
religion as a way to achieve nonreligious goals, “to provide security and solace,
sociability and distraction, status and self-justification. The embraced creed is
lightly held or else selectively shaped to fit [nonreligious] needs” (Allport & Ross,
1967, p. 434). People high in extrinsic otientation are hypothesized to pay litde
attention to religious teachings and so to accept and express their societies’ prejudices
even when those prejudices run counter to their religions’ teachings (Duck &
Hunsberger, 1999). The first two sections of Table 7.4 present some sample ques~
tionnaire items used to assess intrinsic and extrinsic religious orientation, Research
findings have generally supported the hypothesis that an extrinsic orientation is
positively related to prejudice, although the correlations are often small. The results
for intrinsic religiosity are less clear, but studies have generally found either a small
negative or no correlation for racial prejudice (Batson et al., 1993) but a positive
correlation for anti-gay prejudice (Whitley, in press).
The findings that sincere religious belief, defined in tenns of an intrinsic religious
orientation, was negatively correlated with prejudice were nl0re satisfying
258 CHAPTER 7
TAB L E 7.4 Sample Questionnaire Items Used to Assess
Religious Orientations
Intrinsica
It is important to me to spend periods of time in private religious thoughts
and meditation.
I try hard to carry my religion over into all my other dealings in life.
Quite often I have been keenly aware of the presence of God or the Divine Being.
Religion is especially important to me because it answers questions about the
meaning of life.
Extrinsica
Although I believe in my religion, I feel there are many more important things in life.
The church is most important as a place to formulate good social relationships.
I pray chiefly because I have been taught to pray.
Occasionally, I find it necessary to compromise my religious beliefs in order to protect
my social and economic well-being.
Questb
As I grow and change, I expect my religion also to grow and change.
It might be said that I value my religious doubts and uncertainties.
I was not very interested in religion until I began to ask questions about the meaning
and purpose of my life.
Questions are far more central to my religious experience than are answers.
FundamentalismC
God has given mankind a complete, unfailing guide to happiness and salvation, which
mus.t be totally followed.
The long-established traditions in religion show the best way to honor and serve God,
and should never be compromised.
Whenever science and sacred scripture conflict, science must be wrong.
To lead the best, most meaningful life, one must belong to the one, true religion.
“Allport and Ross (1967), reproduced in Batson, Schoenrade, and Ventis (1993, p. 162).
bBatson, Schoenrade, and Ventis (1993, p. 170).
CAltemeyer (1996, pp. 158-159).
to researchers than the finding that religious involvement was positively correlated
with prejudice. However, Batson and colleagues became concerned that intrinsically
religious people might not really be low in prejudice, but instead were simply
motivated not to appear prejudiced (Batson. Flink, Sclioenrade. Fultz. & Pych, r1986). Tliat is. because their religions tell them they should be unprejudiced, intrinsically
religious people give socially desirable-that is, unprejudiced-responses
on self-report measures. As Daniel Batson and Christoplier Burris (1994) put it.
— “every major [religious] denomination is on record opposing racial prejudice and
discrimination. Yet we are given pause when we see a headline that reads, ‘Sunday
morning at 11 remains most segregated hour of the week’ (Atlanta Constitution,
August 9. 1987, p. II-A). We suspect that in many cases the intrinsic believer.
INDIVIDUAL DIFFERENCES AND PREJUDICE 259
attending to the practice of the religious community as well as the preaching, is
~eamingl a very different, more pharisaical norm: The truly religious can’t look
racist” (p. 167).
Batson and his colleagues (1986) tested this hypothesis in an experiment in
which White research participants were led to believe that they and another student
would watch and evaluate a short movie in one of two two-person
“theaters.” When participants arrived at the theaters, they found a Black student
waiting in one and a White student waiting in the other. There were two experimental
conditions. In the overt prejudice condition, the same movie was being
shown in each theater, so choosing to sit with the White confederate might
make parricipants appear to be prejudiced, because the race of the other person
was the only factor that differentiated the two theaters. In the covert prejudice
condition, different movies were being shown in each theater, so choosing
to sit with the White confederate could be attributed to a factor other than
prejudice-the movie. Batson and his colleagues reasoned that if social desirability
influenced the racial attitudes of internally religious people, they would
choose to sit with the Black person in the overt prejudice condition as a way
of demonstrating their lack of prejudice, but would choose to sit with the
White person when that choice could be attributed to movie preference rather
than prejudice. A lack of prejudice would be indicated if an equal number of
participants chose to sit with the Black student and the White student. The
researchers’ results partially supported their expectation: In the overt prejudice
condition, 75 percent of the intrinsically religious participants chose to sit with
the Black student compared to 46 percent of the intrinsically religious participants
in the covert prejudice condition. Note that although the intrinsically
religious participants exhibited a social desirability response bias by favoring the
Black student over the White student, they made unprejudiced choices in the
covert condition, sitting with the Black and White students at about the same
rate. Thus, although intrinsically religious people do appear to be influenced by
social desirability concerns, they also appear to be unprejudiced, at least in regard
to race. Contrary to the theory underlying the concept of religious orientation,
extrinsic religiosity was unrelated to prejudice in either study.
Quest Orientation. Based on his study of theology and the results of his psychological
research, Batson (1976) proposed a third type of religious orientation,
which he named quest. Quest reflects a view of religiosity as a search, or quest,’- \
for answers to questions about the meaning of life. “An individual who approaches
religion in this way recognizes that he or she does not know, and probably never
will know, the final truth about such matters. Still, the questions are deemed
important and, however tentative and subject to change, answers are sought”
(Batson & Burris, 1994, p. 157). The third section of Table 7.4 shows some sample
questionnaire items used to assess quest orientation. Quest orientation is only
minimally correlated with intrinsic and extrinsic orientation (for example,
Wilkinson, 2004) and so constitutes a third dimension of religious orientation
that Batson (1976) described as a “more … flexible type of religiosity than the
other two” (p. 207).
260 CHAPTER 7
Because quest orientation reflects an open-mindedness and willingness to
change that would include tolerance for members of other social groups, researchers
have hypothesized that it would be negatively correlated with prejudice
(Spilka et al., 2003). With few exceptions, research has supported that hypothesis
(Whitley, in press). In addition, the relationship between quest and low prejudice
seems to be unaffected by social desirability concerns. For example, in Batson
and colleagues’ (1986) “movie theater” study described earlier, about half of research
participants high on quest orientation chose to sit with the Black student
in both the overt (44 percent) and covert (54 percent) prejudice conditions, indicating
a lack of prejudice. Thus, quest initially appeared to be the source of
“universal love and compassion” sought by psychology of religion researchers
(Batson, Floyd, Meyer, & Winner, 1999).
[Proscribed versus Permitted Prejudices. Theorists have pointed out that
although most religions teach tolerance toward outgroups, some outgroups may
‘be tolerated more than others (Duck & Hunsberger, 1999). That is, although
some religions proscribe (that is, forbid) some forn1S of prejudice, such as racism,
they may at the same time pennit prejudice against people, such as lesbians and
gay men, who are perceived to violate the religion’s values. Because intrinsically
e-;,eligious people believe strongly in their religions’ teaching-; (for example,
Wilkinson, 2004), researchers hypothesized that they would follow their religions’
teachings regarding proscribed and pennitted prejudices. That is, researchers expected
that intrinsic religiosity would be negatively correlated with proscribed
prejudices and positively correlated with pennitted prejudices. In contrast, researchers
hypothesized that quest orientation would be negatively correlated with
both types of prejudice because of the generalized tolerance it engenders (Batson
et aI., 1993). Extrinsic religiosity has not received much attention in this research
because extrinsically religious people are hypothesized to be more influenced by
societal norms than by religious nonns (Herek, 1987). Because societal nonns
mayor may not coincide with religious norms, it is not possible to fonnulate clear
hypotheses about the influence of religious norms on extrinsically religious people.
One approach has been to examine the correlations of religious orientations
to proscribed and pennitted prejudices, using racism as the proscribed prejudice
and anti-gay attitudes as the pennitted prejudice. The correlations that have
been found for the relationship of intrinsic and quest orientations with both forms
of prejudice have been small. Nonetheless, as hypothesized, researchers have found
intrinsic religiosity to be negatively correlated with racism and positively correlated
with anti-gay attitudes and quest to be negatively correlated with both forms of
prejudice (Whitley, in press).
Another approach to testing the hypotheses is experimental, studying people’s
reactions to individuals who are presented to them as violating or not violating
their values. For example Batson and his colleagues (1999) gave college student
research participants who were high on intrinsic religiosity the opportunity to
help another student earn son1e money. The other student was portrayed as either
heterosexual and needing the money to visit his or her grandparents, as gay and
needing the money to visit his or her grandparents, or as gay and needing the
INDIVIDUAL DIFFERENCES AND PREJUDICE 261
money to attend a gay rights rally. Although levels of helping were high in all
three conditions, participants were more likely to help the student portrayed as
heterosexual than the student portrayed as gay, regardless of why the student
needed help. Thus, intrinsically religious students were reluctant to help someone
who violated their religious values even when the help would promote a valueconsistent
behavior (visiting grandparents) rather than value-violating behavior
(attending a gay rights rally).
In contrast to these results for intrinsic religiosity, Batson and his colleagues
(1999) found that participants high on quest religiosity were equally willing to
help the other student regardless of experimental condition. Thus, students high
on quest were willing to help S0111eone who violated their religious values even
when the help would promote a value-violating behavior. However, Batson and
colleagues (Batson et al., 1999; Batson, Eidelman, Higley & Russell, 2001) noted
that people high on quest might value the openness and tolerance that characterizes
their approach to religion more than any particular religious doctrine, such
as the prohibition against homosexuality. People high on quest, therefore might
be tolerant of people who violate religious principles but might dislike prejudiced
people. To test this possibility, Batson and his colleagues (2001) conducted an
experiment similar to the one just described, except that the student the participants
could help was either tolerant of homosexuality and needed money to visit
his or her grandparents, intolerant of homosexuality and needed money to visit
his or her grandparents, or intolerant of homosexuality and needed money to
attend an anti-gay-rights rally. The researchers found that although participants
high on quest were equally willing to help both the tolerant and intolerant student
visit grandparents, almost none of them were willing to help the intolerant
student to attend the anti-gay-rights rally. That is, people high on quest were
tolerant of someone whose religious beliefS were different than their own, but
would not support behavior that was contrary to their beliefs.
Although these findings shed a positive light on people high on quest, Jerry
Goldfiied and Maureen Miner (2002) proposed that this tolerance may be limited.
They suggested that whereas people high on quest may be tolerant of attitudes
that differed from their own, they may not be tolerant of a religious style
that ran contrary to theirs, such as fundamentalism. Using a research design similar
to Batson and colleagues (2001), Goldfried and Miner found that people
high on quest were unwilling to help a person who expressed a fundamentalist
religious orientation even when the help would not promote fundamentalist
religious goals. Thus, people high on quest appear to be tolerant of prejudiced
people (Batson et aI., 2001), but not of intolerant behavior or people whose religious
style is inconsistent with their own open-minded orientation. Thus, there
appears to be no universally tolerant religious orientation: Intrinsic and quest orientation
are each related to some form of prejudice.
Religious Fundamentalism. Religious fundantentalism is “the belief that\
there is one set of religious teachings that clearly contain the fundamental, basic,
intrinsic, essential, inerrant truth about humanity and deity; that this essential
truth is fundamentally opposed by forces of evil which must be vigorously
262 CHAPTER7
I fought; [and] that this truth must be followed today according to the fundamen__
tal, unchangeable practices of the past” (Altemeyer & Hunsberger, 1992, p. 118).
Religious fundamentalists also are committed to using their belief system as a
guide for understanding and interacting with the secular world (Kirkpatrick,
Hood, & Hartz, 1991). Fundamentalist movements sharing these characteristics
are found among Christians, Jews, and Muslims (Armstrong, 2000). The last section
of Table 7.4 shows some sample questionnaire items used to assess religious
fundamentalism. Not surprisingly, given its nature, fundamentalism is highly correlated
with an intrinsic religious orientation and somewhat negatively correlated
with quest orientation (for example, Rowatt & Franklin, 2004). These correlations
reflect fundamentalists’ adherence to religion and to living their religion in
their everyday lives on the one hand and, on the other, the contrast between
questers’ search for answers to theological question and fundamentalists’ confidence
that their religion already provides those answers.
[
Research has found religious fundamentalism to be consistently associated
with prejudice (Spilka et aI., 2003). For example, in a review of studies published
between 1989 and 2006, Whirley (in press) found that fundamentalism had an
average correlation of r = .44 with negative attitudes toward homosexuality
and an average correlation of r = .18 with negative racial attitudes. Note that,
as with intrinsic religious orientation, the correlation is stronger f<?r a pennitted
prejudice (against homosexuality) than for a proscribed prejudice (racism).
fHowever, whereas intrinsic orientation has a negative correlation with proscribed
L~rejudice, fundamentalism has a positive, albeit small, correlation. In addition,
Aubyn Fulton, Richard Gorsuch, and Elizabeth Maynard (1999) concluded that
“the homosexual antipathy of fundamentalism is in excess of what is required by
their [sic] religious ideology” (p. 20). They came to this conclusion based on two
~ of their research findings. First, although fundamentalists’ religious values require
them to reject homosexuality on moral grounds, they did so on nonmoral
grounds as well, thereby going beyond the requirements of their religion.
Second, Fulton and colleagues (1999) found that fundamentalists expressed prejudice
against celibate as well as sexually active gay men, even though the fonner
group “are not in violation of the perceived biblical injunctions [against homosexual
behavior]” (p. 20).
Conclusions. Given the complexity of the relationship between religion and
prejudice, what can we conclude? Three factors stand out. First, almost all religions
teach acceptance and tolerance of all people, including people belonging to
different racial and ethnic groups. Second, in practice, this acceptance and tolerance
can be limited to those who are perceived to share one’s religious values;
prejudice may be pennitted against those who are perceived to violate those
values. Note, however, that pennitting a prejudice is not the same as requiring it.
That is, people are allowed to adhere to pennitted prejudices, but are not required
to do so; as a result, people nlay or nuy not exhibit a pennitted prejudice based on
other factors that influence their beliefs. Finally, it is essential to bear in mind that
all the data relating religiosity to prejudice are correlational, so one should not
come to the conclusion that religion causes prejudice. Although that might he
INDIVIDUAL DIFFERENCES AND PREJUDICE 263
true in some cases, in other cases people might be using religious doctrine as a
justification for their preexisting prejudices. As Allport (1954) noted more than
50 years ago, “Piety may … be a convenient mask for prejudices which .,. have
nothing to do with religion” (p. 447).
Political Orientation
Political orientation is one of the most controversial topics addressed by those
who study the psychology of prejudice, The controversy arises from research
that has consistently found a correlation between endorsement of conservative
political beliefs and prejudice (for example, Jones, 2002). As a result, some writers
have objected that conservatives have become what might be called the
“designated villains” of prejudice. For example, Paul Snidennan and Philip
Tetlock (1986) have suggested that the typical portrayal is that “Racists … are
by definition conservatives; and conservatives, again by definition, are racists”
(p. 181). The symbolic prejudice approach (see Chapter 6) has been particularly
singled out for criticism in this regard because it defines prejudice partly in tenus
of some of the traditional American values that conservatives endorse (see, for
example, Tetlock, 1994; for a reply, see Sears, 1994).
In this section, we examine some of the recent research on the relation
of political orientation to prejudice, looking first at prejudice itself and then at
attitudes toward social policies, such as affirmative action, that are intended to
relieve some of the effects of prejudice and discrimination, While reading this
section, it is important to bear in mind that there has been little consensus on
how to define liberal and consetvative) the two key terms of political orientation,
either conceptually or operationally (Knight, 1999). As a result, direct compatison
of the results of different studies can be difficult.
Conservatism and Prejudice. As was the case for older research, more receniJstudies
~av~ found a relationship ~etween, en~orsement of conservative beli~fs __ and prejudICe (for example, Fedenco & Sldamus, 2002). However, as Ducbtt j
(1994) noted, the more important question is not whether a relationship exists
between conservatism and prejudice, but why it exists. Two principal explanations
have been proposed, both rooted in concepts we discussed earlier in this
chapter. One explanation draws on the concepts of social dominance orientation'(
SDO) and right-wing authoritarianism (R WA). Endorsement of conservative beliefs
is correlated with both SDO (for example, Whitley & Lee, 2000), which
are themselves related to prejudice. Together, R WA and SDO account for a
large proportion of variance in conservatism (Van Hiel & Mervielde, 2002) and
when SDO and R W A are controlled, the correlation between conservatism and
prejudice is greatly reduced (for example, Federico & Sidanius, 2002). From the
perspective of social dominance theory, these results suggest that prejudice is really
caused by SDO and R W A rather than a conservative belief system: The
correlation between conservatism and prejudice arises because conservative beliefs
constitute one form of legitimiZing myths that people high in SDO can
use to justify their prejudice. That is, conservatism does not cause prejudice;
264 CHAPTER 7
rather, some prejudiced people use the conservative belief system as a llleans of
justifYing their prejudices.
[;
Another explanation for the relationship between conservatism and prejudice
draws on the attribution-value Inodel of prejudice. Researchers have found that,
compared to liberals, conservatives are more likely to see people as being responsible
for negative outcomes they experience, such as poverty and unemployment
(for a review, see Skitka, Mullen, Griffin, Hutchinson, & Chamberlin, 2002).
Therefore, when one group experiences a negative outcome, such as unemployment,
to a greater degree than another group, conservatives are likely to attribute
the outcome to a factor under group members’ control, such as laziness (especially
if laziness is part of the group stereotype), In contrast, liberals are more likely to
attribute the outcome to factors beyond individuals’ control, such as poor
economic conditions. These different perceptions make conservatives more likely
to be prejudiced because, in terms of the attribution-value model, they are more
likely to perceive others as violating an important social value (hard work in our
example) and dislike them for it.
This principle is illustrated by research conducted by Alan Lambert and
Alison Chasteen (1997), who examined prejudice against African Americans
and older adults. Lambert and Chasteen chose these groups because, although
both are perceived to be economically disadvantaged, Mrican Anlericans are
stereotypically blamed for their economic situation because they are seen as violating
the work ethic, but older people are not blamed for their situation because
they are seen as victims of circumstance. The researchers hypothesized that liberalism
would be correlated with positive attitudes toward both groups because
liberals tend to attribute disadvantage to situational factors regardless of value
issues. In contrast, they hypothesized that conservatism would be correlated
with negative attitudes toward African Americans because they are perceived as
value violators, but with positive attitudes toward older people because they are
not. Lambert and Chasteen’s research is unusual in that they assessed liberalism
and conservatism separately, so each research participant received a score on each
ideological dimension. That is, rather than assuming that people are either liberal
or conservative, they assumed that people can have a mixture of liberal and conservative
beliefs. This approach allowed them to examine individual differences
in both liberalism and conservatism. Lambert and Chasteen’s results supported
their hypotheses: liberalism was correlated with positive attitudes toward both
Mrican Alnericans and older adults, whereas conservatisln was correlated with
negative attitudes toward African Americans but with positive attitudes toward
older people.
If at least some of the relation betvveen conservatism and prejudice comes
from attributional differences betvveen liberals and conselvatives, where do these
differences come from? Linda Skitka and her colleagues (2002) found that the
answer lies in liberalism rather than conservatism. Using an approach that classified
people as either liberal or conservative, they found that both liberals and
conservatives initially attributed responsibility for negative outcom_es to the person
experiencing the outcome. They had expected this finding because personal
responsibility is an American value that both liberals and conservatives have
INDIVIDUAL DIFFERENCES AND PREJUDICE 265
internalized. However, liberals were more likely than conservatives to change
their attribution to a situational one (that is, one that fit their value system)
when given an opportunity to elaborate on their responses. Because an attribution
of personal responsibility is consistent -with conservative values, conservatives
have no need to change their attribution. Therefore, what is typically viewed as
the conservative reaction-attributions of personal responsibility for value-violating
outcomes that lead to blame that in tum lead to prejudice—is the “default option”
for both liberals and conservatives.
Liberalism and Prejudice. John Dovidio and Samuel Gaertner (1998) have I
suggested that both liberals and conservatives can be prejudiced, but that prejudice
takes different forms in the two groups. Conservatives, they proposed, I
exhibit modern-symbolic prejudice whereas liberals exhibit aversive prejudic,
(see Chapter 6). Paul Nail, Helen Harton, and Brian Decker (2003) tested this
possibility by having White research participants who had classified themselves as
liberal, moderate, or conservative read a summary of a legal case in which a
police officer had assaulted a motorist. In one experimental condition the police
officer was White and the motorist was Black; in the other condition, the police
officer was Black and the motorist was White. In both conditions, the officer was
acquitted of assault charges in state court despite strong evidence against him (for
example, the assault had been videotaped by a witness) but later was found guilty
of violating the motorist’s civil rights in federal court. Mter reading the case, the
participants rated the extent to which they thought the police officer was being
exposed to double jeopardy; that is, being tried twice for the same offense.
Nail and his colleagues (2003) hypothesized that if modem-symbolic prejudice
is associated with conservatism, then conservatives would rate double jeopardy
exposure as being lower for the Black police officer because doing so would
allow them to express their prejudice subtly, in that the double-jeopardy rating
does not directly criticize the Black officer. In contrast, they hypothesized that if
liberalism is associated with aversive prejudice, liberals would rate double jeopardy
as higher for the Black police officer because they would feel guilty over
their residual negative racial feelings and overconlpensate for their guilt by being
more sympathetic to the Black officer. The researchers proposed no hypotheses
concerning moderates because neither theory of prejudice addresses that situation.
Figure 7.2 illustrates the results of the research. As predicted, liberals favored
the Black police officer and conservatives favored the White police officer; that
is, both liberals and conservatives were prejudiced but in different ways. In contrast,
moderates, who had no political ideology that might bias their responses,
treated both police officers equally.
Social Policy Attitudes. Generally, people with conservative political beliefsJ
hold more negative attitudes toward social policies intended to increase intergroup
equality than do people with liberal political beliefs (for example, Sidanius, Singh,
Hetts, & Federico, 2000). Perhaps the best known and most controversial of these
programs is affinnative action. Three explanations can be offered to explain why
conservatives are more opposed to affimutive action than are liberals. One is that
266 CHAPTER 7
II Black police officer/White motorist
White police otficer/White motorist
Liberal
(Aversive)
Moderate Conservative
(Modern-symbolic)
Political orientation
(presumed prejudice type) of participant
FIG U R E 7.2 Pro~Black Bias in Aversive Prejudice
Politically liberal participants (presumably experiencing aversive prejudice) rated the Black police officer to be
experiencing double jeopardy to a greater extent than the White police officer, politically conservative participants
(presumably experiencing modern-symbolic prejudice) showed the opposite pattern of response, while
politically moderate participants made similar ratings for Black and White officers.
SOURCE: Adapted from Nail, Harton, and Decker (2003, Figures 1, 2, and 3, pp. 758, 760, and 761),
because conservatism is correlated with prejudice, conservatives oppose such programs
because they are prejudiced. The second explanation parallels the first,
holding that because conservatism is correlated with SDO, conservatives oppose
the programs because they are anti-egalitarian. The third explanation has been
variously labeled principled conservatism (Sidanius, Pratto, & Bobo, 1996), principled
objections (Federico & Sidanius, 2002), and principled politics (Sidanius
et al., 2000). This hypothesis proposes that conservatives oppose affinnative action
because they sincerely believe in certain principles-such as fairness, individual
merit, and minimal government-that they see as being inherently inconsistent
with progra111S that favor one group over another, especially when these programs
are mandated or carried out by the government. While reading our summary of
the research evidence bearing on these hypotheses, keep in mind that interpretation
of the research is cOll1_plicated by the facts that the tern1 affirmative action can
have many meanings and that people’s attitudes toward affinnative action vary as a
function of the meaning they apply to it (see Chapter 14). For example, although
most people think affirmative action nleans giving preferential treatment to
INDIVIDUAL DIFFERENCES AND PREJUDICE 267
women and 111embers of minority groups, such an approach is only legal if an
organization is making up for past discrimination. In its most accurate sense, affirmative
action means establishing policies and procedures that ensure equal opportunity
for women and members of minority groups (Crosby, Iyer, Clayton, &
Downing, 2003).
The prejudice explanation of conservative opposition to affinnative action
implies that if levels of prejudice are controlled, there should no longer be a relationship
between conservatism and opposition. In this context controlling means
using statistical procedures that show what a relationship between two variables,
such as political orientation and opposition to affinnative action, would look like
if everyone had similar scores on a third variable that is correlated with the other
two, such as prejudice. In this case, controlling for prejudice means asking what
the correlation between political orientation and opposition to affinnative action
would look like if both liberals and conservatives had similar scores on a measure
of prejudice. The results of such research show that when prejudice is controlled,
there is only a small reduction in the relationship between conservatism and opposition
to affinnative action (for example, Federico & Sidanius, 2002). These
results indicate that there is more to the relationship between conservatism and
opposition to affinnative action than prejudice.
The anti-egalitarianism explanation of conservative opposition to affinnative
action implies that if SDO is controlled, there should no longer be a relationship
between conservatism and opposition. However, as with prejudice, controlling for
SDO reduces the conservatism-opposition relationship only slighdy (for example,
Federico & Sidanius, 2002). Therefore, conservative opposition to affirmation
action cannot be explained entirely in tenus of anti-egalitarianisill.
The principled objections explanation is more difficult to evaluate than the
other two because a number of principles are at issue and no research has systematically
investigated all of them. We focus on one of those principles, fairness.
One conservative objection to affirmative action is that it is inherently unfair because
it favors one group over another. One implication of this position is that
conservatives should be less opposed to affinnative action programs that do not
violate the principle of fairness or merit (such as by offering the same professional
development programs to members of both majority and minority groups) than
to those that do. Research shows that people who believe in the principle that
rewards should be based on qualifications (the merit principle) also support such
equal treatment programs but do not support programs that give members of one
group preference over members of other groups (Eobocel, Son Hing, Davey,
Stanley, & Zanna, 1998). People who support the merit principle are also less
likely to oppose preferential treatment progranls that are instituted as a remedy
for discriminarion (Son Hing, Bobocel, & Zanna, 2002). Because discrimination
is itself an unfair impediment to application of the merit principle, in such a case
affirmative action would promote fairness.
One complication to fairness as a principled objection to affirmative action is
that perceptions of fairness are themselves related to prejudice and SDO:
Prejudiced people and those high in SDO are more likely to endorse unfairness
as an objection to affirmative action than are less prejudiced people and those
268 CHAPTER 7
low in SDO (Federico & Sidamus, 2002), perhaps as a means of justifYing their
prejudice. A second complication is that the fairness objection should apply
equally to all groups, but conservatives are more opposed to affirmative action for
Mrican Americans than for women (for example, Reyna, Henry, Korfinacher, &
Tucker, 2005). Christine Reyna and her colleagues (2005) found that this difference
was related to an aspect of fairness-deservingness. Women are seen as
lnore deserving of the help provided by affinnative action, perhaps because of
the laziness component of the Black stereotype. Reyna and her colleagues
found that when attributions of deservingness were controlled, there was no
difference in approval of race- and gender-based affinnative action programs.
These results also suggest that fairness as a principled objection is somewhat
contaminated by prejudice.
Conclusions. What can we conclude about the relationship of political orientation
to prejudice? First, the research does indicate a moderate correlation between
conservatism and prejudice, but the research also indicates that SDO and R W A
can account for the relationship. That is, some conservatives are prejudiced and
some are not, and those who are prejudiced tend to be high on SDO or R WA.
Second, liberals as well as conservatives can be prejudiced, but it appears that
liberals are more likely to exhibit aversive prejudice and that conservatives are
more likely to exhibit modem-symbolic prejudice. Finally, some evidence shows
that the objections many conservatives make to programs such as affirmative
action are based on principle rather than prejudice, However, prejudiced people
may use those principles to justify objections that actually stem from prejudice.
SUMMARY
Individual difIerence researchers study the ways in which people’s personal characteristics
relate to other variables such as prejudice. Two individual difference variables
that have been consistently linked with prejudice are right-wing authoritarianism
(RWA) and sodal dominance orientation (SDO). The study of authoritarianism as a
factor in prejudice began after World War II. The research initially focused on
what was called the authoritarian personality) a personality type that was postulated
to be especially susceptible to unthinking obedience to authority and to prejudice.
Recent research has focused on R W A, a refined version of the original concept.
People high in R W A are prejudiced against a wide variety of groups, especially
groups that are perceived to violate traditional values and groups that authority
figures condemn. A number of psychological characteristics may predispose
people high in R W A to prejudice, including mental inflexibility, a disinterest in
experiencing new things, a perception of the world as a dangerous place, and a
tendency to organize their worldviews in tenns of ingroups and outgroups.
SDO is a belief system that leads people to want their ingroup to dominate in
society and be superior to outgroups, It is composed of group-based dominance,
the belief than one’s ingroup should dominate in society, and opposition to
INDIVIDUAL DIFFERENCES AND PREJUDICE 269
equality, the belief that societies should be structured so that one group dominates
others. Generally, members of groups that hold more power in a society exhibit
higher levels of SDO. People high in SDO are prejudiced against a variety of
groups, especially those that challenge the legitimacy of social inequality. Psychological
characteristics that might predispose people high in SDO to prejudice
include seeing the world in competitive tenns, belief that other groups’ successes
necessarily come at their ingroup’s expense, and low empathy. People high in
SDO justifY their prejudices with legitimizing myths, or belief systems, such as
group stereotypes that portray outgroups as inferior to the ingroup. Although
SDO may appear to be similar to R WA, the two concepts differ in a number of
ways. For example, SDO enlphasizes relations between ingroups and outgroups
whereas R W A emphasizes obedience to ingroup authority; scores on measures of
SDO and R WA are only slightly correlated; and the two constructs have different
patterns of relationships to some fonus of prejudice.
The self represents our awareness of ourselves as living beings who interact
with the world and the people in it; self-esteem refers to a person’s selfevaluation
as positive or negative. Theorists have proposed that prejudice can
both enhance and protect self-esteem. Prejudice enhances self-esteem by providing
people with groups they can feel superior to. Although one might expect
that it would be people low in self-esteem who are more prejudiced, research
has found a small positive correlation between self-esteem and prejudice. Three
explanations have been proposed for why it has been high rather than low selfesteem
that has been linked to prejudice. One is that people both low and high in
self-esteem are prejudiced, but in different ways; however, because researchers
have not focused on the ways in which low self-esteem people express prejudice,
it looks like people high in self-esteem are the more prejudiced. The second explanation
that has been offered is that the relationship between low self-esteem
and prejudice is indirect; that is, that low self-esteem results in behaviors, such as
avoidance of intergroup contact, that can lead to prejudice. The third explanation
is that there are two kinds of high self-esteem, secure and defensive; people with
defensive high self-esteem do not really value themselves and so bolster their selflinages
by looking down on others. Prejudice protects self-esteem by deflecting
criticism: If the person making the criticism is a member of a devalued group,
the criticism can be shrugged off as meaningless because members of that group
are not qualified to make valid judgments. Thus, researchers have found that being
the recipient of criticism increases prejudice. Conversely, if a person’s self-esteem
is reinforced, the person will exhibit less prejudice.
Values are beliefs people hold concerning the relative itnportance of the goals
they aspire to achieve in life and the types of outcomes they should try to avoid.
Two value orientations have been related to prejudice. Individualism emphasizes
the importance of self-reliance, and egalitarianism emphasizes the importance of all
people being treated equally and fairly. The link between individualism and prejudice
is group stereotypes: Groups that are stereotyped as behaving in ways that
violate the principles of individualism are viewed negatively. In contrast, egalitarianism
appears to be negatively correlated with all fonus of prejudice. Egalitarianism
may have its effects by suppressing stereotype activation.
270 CHAPTER 7
Although some value orientations appear to be directly related to prejudice,
prejudice is also related to the perception that OlltgroUpS’ value systems differ
from one’s own: Because values guide judgments of what is good or bad, holding
different values implies a lack of goodness in the outgroups. Research has
shown that people generally believe that members of outgroups hold values
that differ from their own, and research has also shown that perceptions of value
difference are related to prejudice. One explanation for the values-prejudice relationship
comes from terror management theory, which holds that awareness of
one’s mortality increases one’s adherence to the ingroup’s cultural values; people
who are aware of their 1110rtality express prejudice against groups they see as
challenging those views as a way of deflecting that challenge. The attributionvalue
11lodel holds that groups that are seen as violating values are disliked
because they are seen as choosing to violate those values, and so are responsible
for their negative (that is, value-violating) behavior. In general, people who are
seen as responsible for their negative behaviors and outcon”les are liked less than
people who are seen as not responsible, their behaviors and outcomes being due
to factors they cannot control, such as biology or economic circumstances.
Social ideologies are sets of attitudes and beliefS that predispose people to
view the world in certain ways and to respond in ways consistent with those
viewpoints. Two important social ideologies that have been studied in relation
to prejudice are religion and political orientation. Early research on the relationship
between religion and prejudice focused on religious involvement and found
that involven”lent was positively correlated ¥lith prejudice. Researchers then
examined ways of being religious, and found that intrinsic religious orientation,
which views faith as an end in itself, had a small negative correlation with prejudice
whereas extrinsic orientation, which views religion as a means for achieving
other goals, had a small positive correlation with prejudice. However, other research
suggested that some of intrinsic orientation’s negative relation to prejudice
might have resulted from people’s desire to appear unprejudiced rather than
from a true lack of prejudice. A third orientation, quest, views religion as a process
of seeking answers to life’s important questions, and is negatively correlated
with prejudice.
A more recent approach to studying the religion-prejudice relationship has
focused on the contrast between prejudices that religions proscribe, such as racial
prejudice, and prejudices that some religions pennit, such as anti-gay prejudice.
Generally, intrinsic religiosity has been found to have positive correlations with
permitted prejudices and small negative correlations with proscribed prejudices.
In contrast, quest has been found to have negative correlations with both types
of prejudice. However, people high on quest do appear to be prejudiced against
people, such as religious fundamentalists, whose style of religious belief runs
counter to their own. The results for extrinsic orientation have not been as clear:
Although researchers have found it to have a positive correlation with racial prejudice,
it has had no correlation with anti-gay attitudes.
Religious fundamentalism has been found to be positively correlated with
both anti-gay prejudice aud racial prejudice, although the latter correlation is
llluch smaller. As a psychological trait, fundamentalism is related to both R W A
INDIVIDUAL DIFFERENCES AND PREJUDICE 271
and a strong adherence to orthodox religious beliefs. Thus, religious orthodoxy
triggers hostility toward targets of pennitted prejudice that is supplemented by
hostility arising from compliance with religious authority figures’ condemnation
of those groups.
Researchers have generally found a positive correlation between endorsement
of a conservative political orientation and prejudice. One reason for this
relationship may lie in the positive correlation between conservatism and SDO:
With SDO controlled, the conservatism-prejudice correlation is greatly reduced.
From this perspective, conservative beliefs constitute one foml of legitimizing
myths that people high in SDO can use to justify their prejudice. That is, conservatism
does not cause prejudice; rather, some prejudiced people use the
conservative belief system as a means of justifying their prejudices, A second
explanation for the relationship lies in the attribution-value model of prejudice.
Compared to liberals, conservatives are more likely to see people as being responsible
for negative outcomes they experience, such as poverty and unemployment,
Therefore, when one group experiences a negative outcome to a greater
degree than another group, conservatives are likely to attribute the outconle to a
factor under group members’ control. In contrast, liberals are more likely to attribute
the outcome to factors beyond individuals’ control. These different perceptions
make conservatives more likely to be prejudiced because they are more
likely to perceive others as violating an important social value and dislike them
for it. However, the correlation between conservatism and prejudice does not
mean that liberals are necessarily unprejudiced, Some research suggests that
whereas conservatives experience modem-symbolic prejudice, liberals experience
aversive prejudice,
People with conservative political beliefS tend to hold more negative attitudes
toward social policies intended to increase intergroup equality, such as affinnative
action, than do people with liberal political beliefs. Three explanations have been
offered for this difference, Two of those explanations, that the relationship is due
to conservatives’ being more prejudiced and that it is due to conservatives’ being
high on SDO, have not found much research support. The third explanation, that
conservative opposition is rooted in race-neutral political principles, has found
some support. However, there is also evidence that support for at least some of
these principles is correlated with prejudice. It is therefore possible that those principles
are used by prejudiced people to justify objections that stem from prejudice
rather than principle.
SUGGESTED READINGS
Authoritarianism
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D.]., & Sanford, R. N. (1950). TIl’
authoritarian personality. New York: Harper & Row.
Altemeyer, B. (1996), The authoritarian specter. Cambridge, MA: Harvard University Press,
Stone, W. F., Lederer, G., & Christie, R, (Eds.). (1993). Strength and weakness: The
authoritarian personality today. New York: Springer-Verlag,
272 CHAPTER 7
The authoritariatl personality by Adorno and his colleagues is the classic original
exposition of the theory of authoritarianism. The chapters in Stone and
colleagues’ book summarize much of the research that has been conducted on
the theory. Altemeyer’s book summarizes his revision of the original theory and
presents some of his research on his theory.
Social Dominance Orientation
Altemeyer, B. (1998). The other “authoritarian personality,” Advances in Experimental
Social Psychology, 30, 47-92.
Pratto, F., Sidanius,]., & Levin, S. (2006). Social dominance theory and the dynamics
of intergroup relations: Taking stock and looking forward. European Review of Social
Psychology, 17,271-320.
Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory of social hierarchy and
oppression. New York: Cambridge University Press.
Sidanius and Pratto’s book presents the theory underlying the concept of social
dominance orientation (SDO) and much of the research that has been conducted
on it. Pratto and colleagues’ article provides an update on the theory.
Altemeyer’s chapter compares SDO and right-wing authoritarianism.
Self-Esteem
Crocker, J., Blaine, B., & Luhtanen, R. (1993). Prejudice, intergroup behaviour and
self-esteem: Enhancement and protection motives. In M. A. Hogg & D. Abrams
(Eds.), Group motilJation: Social psychological perspectives (pp. 52-67). New York:
Harvester Wheatsheaf.
Crocker and her colleagues provide an overview of theories that relate
self-esteem to prejudice.
Personal Values
Biernat, M., Vescio, T. K., Theno, S. A., & Crandall, C. S. (1996). Values and prejudice:
Toward understanding the impact of American values on outgroup attitudes. In
C. Seligman, J. M. Olson, & M. P. Zanna (Eds.), The psychology <if values
(pp. 153-189). Mahwah, NJ: Erlbaum.
This chapter provides an overview of the relationship of values to prejudice,
including the roles values play in theories of contemporary prejudice.
Terror Management Theory
Greenberg, j., Solomon, S., & Pyszczynski, T. (1997). Terror management theory of selfesteem
and cultural worldviews: Empirical assessments and conceptual refinements.
Advances in Experimental Social Psychology, 29, 61-139.
Pyszczynski, T., Solomon, S., & Greenberg, J. (2003). In the wake <if 9/11: TIle psychology
oj terror. Washington, DC: American Psychological Association.
INDIVIDUAL DIFFERENCES AND PREJUDICE 273
Solomon, S., Greenberg,]., & Pyszczynski, T. (2000). Pride and prejudice: Fear of death
and social behavior. Current Directions ill Psychological Science, 9, 200-204.
The chapter by Greenberg and colleagues provides a detailed explanation of
terror management theory. The article by Solomon and colleagues presents a
succinct overview of the theory focusing on prejudice, and the book by
Pyszczynski and colleagues interprets the September 11, 2001, terrorist attacks
on the United States in light of their theory.
Attribution-Value Model
Crandall, C. S., D’Anello, S., Sakalli, N., Lazarus, E., Wieczorkowska, G., & Feather,
N. T. (2001). An attribution-value model of prejudice: Anti-fat attitudes in six
nations. Personality and Sodal Psychology Bulletin, 27, 30-37.
Weiner, B. (1995). Judgments if responsibility: A foundation for a theory if sodal conduct. New
York: Guilford.
The article by Crandall and his colleagues outlines the attribution-value model.
Weiner’s book presents the theory on which the model is based.
Religion
Batson, CD., Schoenrade, P., & Ventis, W. L. (1993). Religion al1d the individual:
A social~psydlOlogjcal perspective. New York: Oxford University Press.
Spilka, B., Hood, R. W., Jr., Hunsberger, B., & Gorsuch, R. (2003). The psychology
ifreligiol1: An empirical approach (3rd ed.). New York: Guilford.
Batson and colleagues’ Chapter 9 provides an overview of research and theory
on the relations of religion to prejudice, which Spilka and colleagues update in
their Chapter 14.
Political Orientation
Nail, P. R., Harton, H. C, & Decker, B. P. (2003). Political orientation and modern
versus aversive racism: Tests of Dovidio and Gaertner’s (1998) integrated model.
Journal of Personality and Social Psychology, 84, 754-770.
Sidanius, j., Singh, P., Hetts, J. j., & Federico, C. (2000). It’s not affirmative action, it’s
the Blacks: The continuing relevance of race in American politics. In D.O. Sears,
J. Sidanius, & L. Bobo (Eds.), RadaUzed politics: TIle debate about racism in America
(pp. 191-235). Chicago: University of Chicago Press.
Snidennan, P. M., Crosby, G. C., & Howell, W. G. (2000). The politics of face.
In D.O. Sears, J. Sidanius, & L. Bobo (Eds.), Racialized politics: TIle debate about
racism in America (pp. 236-279), Chicago: University of Chicago Press.
Nail and his colleagues present evidence that both liberals and conservatives
are prejudiced, but in different ways, Sidanius and his colleabrues and
Sniderman and his colleagues present different sides of the principled
objections debate.
274 CHAPTER 7
attribution-value model
authoritarian personality
egalitarianism
extrinsic religious
orientation
ideologies
individualism
intrinsic religious
orientation
KEY TERMS
legitimizing myths
quest orientation
religious
fundamentalis111
right-wing
authoritarianism
(RWA)
self-enhancement
self-protection
social dominance
orientation (SDO)
symbolic beliefs
symbolic threats
terror Inanagement
theory
value difference
hypothesis
QUESTIONS FOR REVIEW AND DISCUSSION
1. Explain authoritarianislll as conceptualized by Adorno and his colleagues
(1950) and by Altemeyer (1981). In what ways are those conceptualizations
similar and in what ways do they differ?
2. Describe the characteristics of people high in right-wing authoritarianism
(R W A) that may predispose them to prejudice. How are these characteristics
related to the ones described in Chapter 4 that are associated with a
propensity for stereotyping?
3. Explain the role authority figures play in shaping the prejudices of people
high in RWA.
4. What is social dominance orientation (SDO)? In what ways is it similar to R WA
and it what ways does it differ from R W A?
5. Describe the characteristics of people high in SDO that may predispose them
to prejudice. How are these characteristics related to the ones described in
Chapter 4 that are associated with a propensity for stereotyping?
6. Explain the concept of legitimizing myths as it is used in social dominance
theory.
7. Think back to Chapter 6. Do any of the theories of contemporary prejudice
postulate that people use what social dominance theory calls legitimizing
myths to justifY their prejudices? If so, what are those myths?
8. Explain the distinction between self-esteem enhancement and self-esteem
protection as motives for prejudice. How are these motives related to the
ones described in Chapter 4 that are associated with a propensity for
stereotyping?
9. What are the two hypotheses that can be derived from the self-enhancement
view of the role self-esteem plays in prejudice? What has research on these
INDIVIDUAL DIFFERENCES AND PREJUDICE 275
hypotheses found? What explanations have been offered to account for these
findings?
10. What has research on the relationship between self-esteem protection
mDtives and prejudice found?
11. What are values? Describe the value orientations that have been studied in
relation to prejudice. How is each related to prejudice? What processes have
linked each value orientation to prejudice?
12. Explain how the perception of value dissimilarity can lead to prejudice.
What individual difference variables are related to this process?
13. Describe the terror management theory explanation for the role played by
perceived value dissimilarity in prejudice.
14. Describe the attribution-value explanation for the role played by perceived
value dissimilarity in prejudice. How are perceptions of a group’s naturalness
and entitativity related to this process?
15. What are social ideologies? In what ways do religion and political
orientation fit the definition of an ideology? In what ways do they not fit
that definition?
16. Allport (1954) wrote that religion “makes prejudice and it unmakes
prejudice” (p. 444). What did he mean by that? What light has subsequent
research shed on his statement?
17, Describe the relationship of religious involvement to prejudice. Why were
religious leaders and psychology of religion researchers displeased by these
findings? What new approach did researchers take?
18, Explain the concepts of intrinsic and extrinsic religious orientation, In
theory, how should each be related to prejudice? What has research shown
about how each is related to prejudice?
19, Debate the following proposition: Intrinsically religious people are no less
prejudiced than anyone else; they are just lllore motivated to give socially
desirable responses to questions about prejudice,
20. Explain the concept of quest as a religious orientation, How is it related
to prejudice?
21. What is meant by proscribed and permitted prejudices? We used racial
prejudice as an example of a proscribed prejudice and anti-gay attitudes
as an example of a pemutted prejudice, What other examples can you
think of for each category? Is the concept of proscribed versus permitted
prejudices unique to the religious context or does it apply to society in
general?
22, Describe how each of the three religious orientations is related to proscribed
and permitted prejudices.
23, Define religious fundamentalisnl, How is it related to prejudice? What seem
to be its major psychological components? How might each of these
components contribute to fundamentalism’s relationship to prejudice?
276 CHAPTER 7
24. Allport (1954) wrote that “piety may … be a convenient mask for prejudices
whicb … have nothing to do with religion” (p. 447). What did he mean?
25. Researchers generally find a positive correlation between political
conservatism and prejudice. What explanations have been offered to
account for that relationship?
26. Describe the relationship between political liberalism and prejudice.
27. Researchers generally find a positive correlation between political
conservatism and attitudes toward social policies, such as affirmative
action, that are intended to increase intergroup equality. What explanations
have been offered to account for that relationship? Which explanation do
you think is best supported by the research evidence? Explain the reasons
for your answer.
28. Debate the following proposition: Political conservatism may be a
convenient mask for prejudices that have nothing to do with politics.